With the fantasy baseball season nearly a month in, Dalton Del Don delivers his latest bullpen update.

Fantasy Baseball Bullpen Brief: Will the top closer by ADP bounce back?

Kaleb Johnson could outperform ADP if he lands in one of these three spots

Best Fantasy Football landing spots for Kaleb Johnson in the 2025 NFL Draft

Chris Towers provides an early look at whether the Spring Training hype was real for the biggest ADP risers

Fantasy Baseball: An early check-in on the biggest ADP risers, from Cam Smith to Kristian Campbell

The Mobility Scheme, or Accessible Vehicles and Equipment Scheme as it is now known for those in Scotland, offers an affordable way for disabled people to get around more easily.

Motability Scheme update for people on PIP, ADP and other disability benefits

In the latest sign that the labor market continues to cool, the pay premium for workers who leave their jobs shrank in March to its lowest level since ADP began tracking the data in November 2020.

'Pay premium' for switching jobs hits post-pandemic low

SportsLine simulated the entire MLB season and identified 2025 Fantasy baseball busts that won't live up to their ADP

Fantasy Baseball Rankings 2025: Busts via proven MLB model that called Jordan Montgomery's tough season

Fantasy Baseball Rankings 2025: Busts via proven MLB model that predicted Jordan Montgomery's tough season

The Fantasy Plus Experts Consensus model has identified three starting pitcher bargains after pick 150 in Yahoo drafts.

Yahoo Fantasy Plus identifies 3 of 2025's top pitcher sleepers - how much potential do they offer at ADP?

The new platform has more than 5,000 designers and mentors in a supportive, global community to help designers skill-up and advance their careers. Design and collaboration tool software provider InVision has had a productive year. In November 2020, they . In January, they announced another project, meant to give back to the design community, with the Amazing Design People List (ADPList) a free platform for designers to build on their skills and connect with mentors. reported a gain of 2 million new users InVision believes in empowering designers and empowering companies with “ ” -- which is explained by their Chief Design Evangelist Stephen Gates on the Growth Manifesto podcast. design intelligence They have the distinction of working with all 100 of Fortune 100 companies, so their roster of Design Mentors hail from all over the world and include major brands like Nike, Apple, and Google along with smaller companies and design-focused recruiting firms. The platform has been quietly growing over the past 10 months and in January the platform fully launched with profiles, easier sharing, and more mentors around the globe. The platform now has more than 5,000 designers and mentors in 18 languages from 30+ countries that currently use the platform. I spoke with Jennifer Aldrich, Director of Community Engagement at InVision to learn more about the platform. Jennifer Aldrich, Director of Community Engagement at InVision What led to the creation of ADPList? The digital product design community is integral for organizations forced to accelerate their digital transformation efforts. Yet (that) couldn’t keep designers safe during the that started in March of 2020, when the pandemic truly started to take its toll on the global economy. Designers, particularly junior designers, were losing their jobs in record numbers. mass layoffs Felix Lee in Singapore and James Badour in Ghana, both accomplished designers, found this unacceptable. Their partnership started with the realization that there was no easy way for laid-off designers to list themselves as looking for a job, and no central clearinghouse where designers could connect with other designers to secure employment, mentorship, or even just support during a difficult time. So they put together a public spreadsheet where people could mark themselves as “looking for a job” or “ready to mentor,” and called it the Amazing Design People List (ADPList). It took off around the global design community. That weekend, they decided to hack the first website on Webflow. It struck a nerve, and hundreds of people signed up. Mentoring happened almost immediately, and some of the laid-off designers started finding jobs. But, given the demand, Felix and James needed to scale. James and Felix knew Stephen Gates, head design evangelist at InVision, who helped them apply for funding from the . InVision Design Forward Fund How do you see this new platform working? It’s a simple concept: a platform to connect designers with mentors who help them uplevel skills and connect them to companies with open roles. Designers join and, with the relaunch of the platform as of January 14, 2021, can now build a profile for free. Mentors give back to the community by opening up their calendars (literally, via a Calendly integration) for meaningful 1:1 connections. Recruiters search through profiles and post jobs. Since April, ADP List has grown to 5,000+ participants and arranged dozens of job placements. What role can mentorship play to help designers? For unemployed designers (and frankly, for many professional tracks) there’s a lack of feedback during the recruiting process. When someone doesn’t get an offer, they’re usually left wondering why. What set the winning candidate apart from me? What do I need to work on? These lingering questions are even more critical in a contracting job market suffering through a pandemic. Design, more so than other industries, is a discipline that relies on how a potential hire shares and explains past creative and design work as an indicator of future success. Mentorship, networking, and even friendship improve designers’ enough so that they can land an even better job than what would have been possible before. skills and marketability Functionally, mentorship can happen in myriad ways. It can include everything from portfolio reviews to career guidance, mock interviews, and leadership advice. Mentors help designers find their voice and get their footing in the industry, which is particularly important for fresh graduates. Access to mentorship should be limited to those who gain access to an exclusive club. It should be available to all. ADPList operates on the principle that mentorship is for everyone and needs to be accessible to all. Individual designers must be supported so that design continues to be elevated in business. On an individual level, it’s ADPList’s goal to help designers find exciting, challenging, and worthwhile job opportunities in today’s market, and to give mentors an easy way to give back to the industry that’s given them so much. It is the entire community of designers—including recent boot camp graduates, unemployed designers, and high-level design directors too—who will shape the digital products that the world uses to navigate our every day. It’s critical that the design community is harnessed, supported, and deeply connected, ready and able to build the best and brightest future of our world. What are the parameters for getting on the list? It is not an exclusive group. Quite the opposite. There are no parameters. It takes just a minute to complete a profile, and you’re in. There is a higher standard for mentors, though. The ADPList is only as good as the advice the mentors provide. The criteria for mentorship include elements such as past mentorship and leadership experience and high technical abilities among others. Is this global, or mainly focused on English speaking audiences? It’s a global platform with founders spread across the world. ADPList is in 30+ countries and conducts sessions in 18+ languages. Most of the meetings, which include one partner meeting and two major events per month, have been in English, but as the community expands, the languages used and the countries represented will, too. Language should not be a barrier to community involvement. That’s also one reason why ADPList uses InVision’s virtual whiteboard for event icebreakers: it’s welcoming and easy for anyone to use, regardless of language. Freehand Beginning this year, ADPList will also host Genius Hours, which are dedicated opportunities for mentors to teach other designers skills beyond design. How does the Amazing Design People list fit in with InVision’s larger strategy to cultivate collaboration between designers? The ADPList is the personification of collaboration that InVision knows is instrumental to design. Great digital product design is now so complex that no one person, or even one team, can bring a great product to market single-handedly. Only structured and inclusive collaboration between teams can bring great products to consumers. And such collaboration is only possible when designers feel confident in their discipline and are elevated within their businesses. ADPList raises up the design community and nurtures a vast network of leaders and up-and-comers around the globe. By supporting ADPList, InVision is able to bolster an entirely new community of designers at a time when they need it most. In the future, this group will continue to share learnings, deepen connections, and collaborate. Both the higher-level skills and the connections these experiences foster will build the foundations of strong design-led product teams that make an outsize impact on the products we all use every day. Summary This is my fourth feature of free and freemium online resources for designers and those, like me, who just occasionally do some design.  This is another worthwhile free tool -- one that may have more of an impact than being able to or create . You can register at no cost as a Designer or Mentor with a range of specializations including AV/VR Design, Game Design, Design Writing, Design Ops, Growth Design, Industrial Design, Motion Design, etc. Once you go through the 3-step process to register as a designer (which I did, though that is an exaggeration to the point of distortion on my part), the platform presents you with three Mentors and the option to schedule a session with any of them. Once you create your profile, you can let ADPList continue to match you with Mentors or search them yourself based on Location, Language, Topic, and Areas of Interest. edit TikTok videos Zoom backgrounds I did not set up a mentoring session since I am not a “real” designer -- but one can see how the platform is potentially pretty exciting for designers in early and mid-career (or maybe beyond) that want to connect with other designers and learn more about how they do what they do. I’ve had the good fortune to have a few great mentors in my own career, for writing, marketing, and PR who I know had a real impact on my career. I’m not a designer, but I see ADPList as a simple good for the design community, and FWIW if I were a designer I would definitely check it out.

ADPList Matches Laid-Off Designers with Mentors for Free

Startups of The Year 2024 Winners: Asia

الفوركس الفوركس 2024 الفائزين: آسيا

২০২৪ সালের বিজয়ী উদ্যোক্তা: এশিয়া

Startups des Jahres 2024 Gewinner: Asien

Startups del Año 2024 Ganadores: Asia

Startups nan ane a 2024 genyen: Azi

2024年のスタートアップ 優勝者:アジア

2024-ийн Startups Шагнагч: Ази

Startup Tahun 2024 Pemenang: Asia

Startups do Ano 2024 Vencedores: Ásia

Startups of The Year 2024 Waqtiga: Asia

Стартапи року 2024 Переможці: Азія

Startups of the Year 2024 laureatlar: Asia

Big Tech Loves These 45 System Design Questions—Do You Have the Answers?

Governments around the world are escalating demands for encryption backdoors, claiming they are required for national security needs. Experts warn that these policies create cybersecurity risks for billions of people. Recently, British officials secretly ordered Apple to create a backdoor into their Advanced Data Protection (ADP) service. Apple responded to this by disabling ADP in the UK – choosing to roll back the optional protection for their UK-based customers, to safeguard the cybersecurity of millions of Apple users around the world. The action from the UK Government, enabled by their Investigatory Powers Act, sets a dangerous precedent. If enforced, this may normalize state-mandated vulnerabilities that hackers, hostile nations and authoritarian regimes could exploit. The encrypted messaging app Signal is facing similar demands from some branches of Swedish Government. If enforced, Signal claims they will leave Sweden. The Domino Effect of Backdoor Demands Apple’s decision to withdraw encrypted cloud storage in the UK—rather than comply—highlights the dilemma facing tech companies. Similar laws in Australia and the US, such as the Assistance and Access Act and FISA Section 702, already empower governments to compel decryption, potentially creating backdoor vulnerabilities. Joseph Lorenzo Hall of the Internet Society predicts a “Commonwealth effect,” where one nation’s backdoor demand triggers copycat policies globally. Cybersecurity experts universally condemn backdoors as systemic risks. As the NSA’s 2006 encryption sabotage showed, intentional weaknesses inevitably leak, enabling mass surveillance and cybercrime. The FBI and CISA now urge Americans to adopt end-to-end encryption for protection—even as the US pressures companies for access. There is a clear lack of unity between government agencies and departments on this issue, not to mention other recent contentions between the FBI and Elon’s newly minted Department of Government Efficiency (DOGE). Will Backdoors Ruin Encryption? Whether or not backdoors become pervasive in encryption is at this point, unclear – but we are clearly at a critical juncture, with multiple spats between public and private entities occurring globally. Governments may abandon backdoor mandates, and instead focus on pushing fully homomorphic encryption (FHE), which enables data processing and analysis without decrypting data. This may be the best case scenario based on current technology. Protecting your Privacy in an Era of Surveillance and Uncertain Encryption Until FHE sees widespread adoption (if that day ever comes), individuals and organizations should combine robust encryption measures whenever possible, while maintaining a strict data hygiene process. Data hygiene can be improved for all people and businesses by adopting a strategy to reduce the size of your digital footprint through careful management. Old social media posts, comments, likes and private messages regularly contain sensitive information that could be exposed in data breaches or misused by third parties. Without a well-managed digital footprint, you are exposed to targeted scams, hackers, and mass surveillance. This is where tools like redact.dev offer a practical safeguard. Redact is a toolkit that gives you the ability to clean up, and automatically manage your digital footprint across 30+ platforms – from Twitter / X Likes and Facebook Activity, to your Slack DMs. Get Redact – free for our most popular services, and reduce attack surfaces for hackers and social engineering vulnerabilities. Governments around the world are escalating demands for encryption backdoors, claiming they are required for national security needs. Experts warn that these policies create cybersecurity risks for billions of people. Governments around the world are escalating demands for encryption backdoors, claiming they are required for national security needs. Experts warn that these policies create cybersecurity risks for billions of people. Recently, British officials secretly ordered Apple to create a backdoor into their Advanced Data Protection (ADP) service. Apple responded to this by disabling ADP in the UK – choosing to roll back the optional protection for their UK-based customers, to safeguard the cybersecurity of millions of Apple users around the world. into their Advanced Data Protection (ADP) service. the cybersecurity of millions of Apple users around the world. The action from the UK Government, enabled by their Investigatory Powers Act, sets a dangerous precedent. If enforced, this may normalize state-mandated vulnerabilities that hackers, hostile nations and authoritarian regimes could exploit. hackers, hostile nations and authoritarian regimes could exploit. The encrypted messaging app Signal is facing similar demands from some branches of Swedish Government . If enforced, Signal claims they will leave Sweden. demands from some branches of Swedish Government The Domino Effect of Backdoor Demands The Domino Effect of Backdoor Demands Apple’s decision to withdraw encrypted cloud storage in the UK—rather than comply—highlights the dilemma facing tech companies. Similar laws in Australia and the US, such as the Assistance and Access Act and FISA Section 702 , already empower governments to compel decryption, potentially creating backdoor vulnerabilities. the Assistance and Access Act FISA Section 702 Joseph Lorenzo Hall of the Internet Society predicts a “Commonwealth effect,” where one nation’s backdoor demand triggers copycat policies globally. predicts a “Commonwealth effect,” Cybersecurity experts universally condemn backdoors as systemic risks . As the NSA’s 2006 encryption sabotage showed, intentional weaknesses inevitably leak, enabling mass surveillance and cybercrime. universally condemn backdoors as systemic risks NSA’s 2006 encryption sabotage The FBI and CISA now urge Americans to adopt end-to-end encryption for protection—even as the US pressures companies for access. There is a clear lack of unity between government agencies and departments on this issue, not to mention other recent contentions between the FBI and Elon’s newly minted Department of Government Efficiency (DOGE). adopt end-to-end encryption FBI and Elon’s newly minted Department of Government Efficiency (DOGE). Will Backdoors Ruin Encryption? Will Backdoors Ruin Encryption? Whether or not backdoors become pervasive in encryption is at this point, unclear – but we are clearly at a critical juncture, with multiple spats between public and private entities occurring globally. Governments may abandon backdoor mandates, and instead focus on pushing fully homomorphic encryption (FHE), which enables data processing and analysis without decrypting data. This may be the best case scenario based on current technology. fully homomorphic encryption Protecting your Privacy in an Era of Surveillance and Uncertain Encryption Protecting your Privacy in an Era of Surveillance and Uncertain Encryption Until FHE sees widespread adoption (if that day ever comes), individuals and organizations should combine robust encryption measures whenever possible, while maintaining a strict data hygiene process. Data hygiene can be improved for all people and businesses by adopting a strategy to reduce the size of your digital footprint through careful management. Old social media posts, comments, likes and private messages regularly contain sensitive information that could be exposed in data breaches or misused by third parties. Without a well-managed digital footprint, you are exposed to targeted scams, hackers, and mass surveillance. This is where tools like redact.dev offer a practical safeguard. Redact is a toolkit that gives you the ability to clean up, and automatically manage your digital footprint across 30+ platforms – from Twitter / X Likes and Facebook Activity , to your Slack DMs . Get Redact – free for our most popular services, and reduce attack surfaces for hackers and social engineering vulnerabilities. redact.dev Twitter / X Likes Facebook Activity Slack DMs

Encryption Wars: Governments Want a Backdoor, but Hackers Are Watching

Cifrado ukan ch’axwawinakapa: Gobiernonakax mä qhipa punku munapxi, ukampis Hackers ukanakax uñch’ukipxi

Şifrələmə müharibələri: Hökumətlər arxa qapı istəyir, lakin hakerlər izləyir

Enkripcijski ratovi: Vlade žele backdoor, ali hakeri gledaju

Verschlüsselungskriege: Regierungen wollen eine Hintertür, aber Hacker beobachten

Guerras de cifrado: los gobiernos quieren una puerta trasera, pero los piratas informáticos están vigilando

暗号化戦争: 政府はバックドアを望んでいるが、ハッカーは監視している

Шифрлөө согуштары: Өкмөттөр арткы эшикти каалайт, бирок хакерлер карап турушат

Perang Penyulitan: Kerajaan Mahu Pintu Belakang, tetapi Penggodam Memerhati

Війни шифрування: уряди хочуть бекдор, але хакери спостерігають

خفیہ کاری کی جنگیں: حکومتیں بیک ڈور چاہتی ہیں، لیکن ہیکرز دیکھ رہے ہیں۔

Iimfazwe ezifihliweyo: oorhulumente bafuna i-backdoor, kodwa iiHackers zibukele

Izimpi Zokubethela: Ohulumeni Bafuna I-Backdoor, kepha Abaduni Babukele

The Beginning of My Mentorship Journey I am from Ukraine, a country that has known very hard times. During the times when war could break out at any moment or had already begun, many people lost their jobs and had to look for work elsewhere. This experience made me realize the importance of sharing my knowledge to help others. It was not only about technical knowledge but also about encouraging people and showing them a way out. The first people I helped were my friends, the friends I had in Ukraine, and the friends I had to help leave the country. I founded ADPList, a platform on which I was able to mentor people around the world. I made a decision that changed my life - I decided to spend at least 30 minutes to an hour each day teaching others. Truth be told, this was one of the best decisions I have ever made. The Importance of Mentorship in My Evolution When I became a mentor, I did not just learn how to teach others, it became a process of self-development. Doing software development is one thing, but sharing knowledge is quite another – you have to become a psychologist, a teacher, and a friend. This realization made me grow not only in coding skills but also in communication, patience, and empathy. Mentoring also assisted me in developing my public reputation and personal brand. I also started posting my live sessions, tutorials, and algorithm challenges on my YouTube channel. I grew my TikTok following and was an active contributor on HackerNoon. Also, I spent a lot of time developing and updating open-source projects on GitHub to use real-world coding examples for my mentees. On the way, I met a lot of interesting people, for example, Greg, a developer from Ukraine, who is engaged in the free teaching school Women For The Future. This school teaches Ukrainian women who have become jobless due to the crisis, develops their development skills, assists them in getting ready for interviews, and offers other forms of support like free mental health care. Greg and I even started a TikTok podcast to promote mentorship work. A Recognized Mentor Today, I am proud to be in the Top 1% Mentor Recognition – January 2025 on ADPList! This is very important to me because it shows how much I have influenced other people’s lives and how much I have developed as a person during this process. In mentorship, I have enhanced my skills, developed my communication skills, and formed strong ties with people from different parts of the world. https://adplist.org/community-certifications/top1percent-jan-2025-engineering-7a0a70 Why You Should Begin Mentoring Today Some people do not like to share their knowledge because they think it is a waste of time. But in fact, mentoring is an investment in yourself. It makes you a better person, makes you more skilled, more outgoing, and helps you to meet fantastic people. It is a process of development and exchange of experiences. If you want to change your life, then start mentoring today. You will be better, more experienced, and surrounded by friends and colleagues who will thank you for your guidance. The Beginning of My Mentorship Journey I am from Ukraine, a country that has known very hard times. During the times when war could break out at any moment or had already begun, many people lost their jobs and had to look for work elsewhere. This experience made me realize the importance of sharing my knowledge to help others. It was not only about technical knowledge but also about encouraging people and showing them a way out. The first people I helped were my friends, the friends I had in Ukraine, and the friends I had to help leave the country. war could break sharing my knowledge to help I founded ADPList , a platform on which I was able to mentor people around the world. I made a decision that changed my life - I decided to spend at least 30 minutes to an hour each day teaching others. Truth be told, this was one of the best decisions I have ever made. ADPList least 30 minutes to an hour each day The Importance of Mentorship in My Evolution When I became a mentor, I did not just learn how to teach others, it became a process of self-development. Doing software development is one thing, but sharing knowledge is quite another – you have to become a psychologist, a teacher, and a friend . This realization made me grow not only in coding skills but also in communication, patience, and empathy. Mentoring also assisted me in developing my public reputation and personal brand. I also started posting my live sessions, tutorials, and algorithm challenges on my YouTube channel . psychologist, a teacher, and a friend live sessions, tutorials, and algorithm challenges my YouTube channel I grew my TikTok following and was an active contributor on HackerNoon. Also, I spent a lot of time developing and updating open-source projects on GitHub to use real-world coding examples for my mentees. On the way, I met a lot of interesting people, for example, Greg , a developer from Ukraine, who is engaged in the f ree teaching school Women For The Future . This school teaches Ukrainian women who have become jobless due to the crisis, develops their development skills, assists them in getting ready for interviews, and offers other forms of support like free mental health care. Greg and I even started a TikTok podcast to promote mentorship work. TikTok GitHub Greg ree teaching school Women For The Future TikTok podcast to promote mentorship TikTok podcast to promote mentorship A Recognized Mentor Today, I am proud to be in the Top 1% Mentor Recognition – January 2025 on ADPList! This is very important to me because it shows how much I have influenced other people’s lives and how much I have developed as a person during this process. In mentorship, I have enhanced my skills, developed my communication skills, and formed strong ties with people from different parts of the world. Top 1% Mentor Recognition – January 2025 on ADPList! https://adplist.org/community-certifications/top1percent-jan-2025-engineering-7a0a70 https://adplist.org/community-certifications/top1percent-jan-2025-engineering-7a0a70 Why You Should Begin Mentoring Today Some people do not like to share their knowledge because they think it is a waste of time. But in fact, mentoring is an investment in yourself. It makes you a better person, makes you more skilled, more outgoing, and helps you to meet fantastic people. It is a process of development and exchange of experiences. If you want to change your life, then start mentoring today. You will be better, more experienced, and surrounded by friends and colleagues who will thank you for your guidance.

My Story as a Mentor: How Learning and Teaching Impacted My Life

HackerNoon editorial team has launched this interview series with women in tech to celebrate their achievements and share their struggles. We need more women in technology, and by sharing stories, we can encourage many girls to follow their dreams. Share your story today! Tell us about yourself! Hi! I'm Lutsia, and I work in HR in IT. I've been doing this for over 12 years now, and I've worked with some amazing companies like gamedev, dating, music-tech, and social. Now, I'm the HR director of an international company that's doing well developing cool mobile apps for social media. My job is all about making sure the company has a great HR strategy, taking care of all the employees, and helping the business grow by developing its people. My company's products are doing well (currently, only Tinder has more installs than us) and we're seeing over 6 million installs every month. I am an active mentor in various communities such as Women in Tech, Miranna.io, ADP.list, I have done more than 100 consultations and helped more than 100 people to achieve their goals in IT. Why did you choose this field in the first place? I've always loved studying people and watching talent grow, so it's a perfect fit for me. In HR, it's all about communication, and in the tech industry, it's a bit of a challenge because there are so many introverts who need individual attention. I've got a Master's in Human Resources Management, and I absolutely love what I do. What tech are you most worried about right now and why? I'm really interested in technologies that can predict how long employees will stay with a company (like prediction models for employee tenure) and assessing soft skills. Human psychology is such a complex system, isn't it? I find the standard approaches don't always work, which makes it a constant challenge in the world of people management to assess team effectiveness. What are your hobbies and interests outside of tech? When I'm not doing that, I love nothing more than spending time with my family and friends, or enjoying a good book. I'm a music lover – playing the keyboard and singing are two of my favourite pastimes. I'm also a keen outdoorsy person, and I love nothing more than a good hike! A couple of years ago, I discovered the wonderful world of mentoring, which is a really cool side activity that energises me by interacting with talented people and being useful. Let's talk about breaking the glass ceiling. What were the biggest challenges you faced as a woman in tech, and how did you deal with them? I started my career in the tech industry, in startups and scale-ups, so I had direct contact with founders and was lucky to avoid unpleasant situations with the glass ceiling. But I would like to see more girls among CEOs and founders of companies. I plan to launch my own project one day. Any questionable misogynistic story/situation you faced/handled, and you want to share with the HackerNoon Fam? When I was just starting out in my career, over 12 years ago now, I was working in talent acquisition. I remember how hiring managers would sometimes reject female candidates for developer positions just because they were girls, or because they thought programming was just not a job for women. I remember my first driving instructor always telling me that driving was not for women, but I don't encounter this now. As a HR leader, I try to nip such stories in the bud. What was the biggest setback/failure that you faced, and how did you manage it? My biggest failure was my personal and professional burnout in 2019. I was deeply immersed in the topic and clearly knew the markers of this condition, but unfortunately, I was not able to prevent it in time. However, I learnt my lesson, and I am now much more attentive to my energy levels and have improved my delegating skills even more! What's your biggest achievement that you're really proud of? I have reached a high position just in 7 years - HR director in a fast-growing global technology company, every day I do what I love and develop not only the team, overseeing the unlocking of immense human potential, but also being involved in creating products that make a difference in this world and make life better for millions of people. I have extensive mentoring experience and have helped hundreds of people find their successful path to self-realisation in the industry. I am surrounded by amazingly smart and caring people who create beautiful things. In your opinion, why do we see this huge gender gap in the tech industry, and how can we reduce it? The tech industry is still pretty new, so I think there are some historical and cultural things going on that are holding it back. But I do see a lot of tech companies trying to make things better, which is great. They're putting in place these DEI policies, which is really important. I'm part of a group that supports women in IT, and I think it's a vital mission. I would love to see more support, grants, quotas and investments for start-ups created by women.It would be great to change the media narrative and promote the idea that technology is a universal field where gender doesn't matter. We should be creating content that's focused on engaging women and breaking stereotypes. From a company perspective and as a C-Level, it's important to implement training to overcome unconscious bias. Who is your tech idol? Why? I really look up to Sheryl Sandberg, ex-COO of Meta and author of Lean In: Women, Work, and the Will to Lead. She started the global Lean In movement, which connects women through professional networks, support circles, and educational programs. This movement helps women in IT find like-minded people, develop skills and build careers. I really admire how she's so open about the personal struggles she's had, including the loss of her husband. It's so inspiring to see how she's managed to be so successful and still remain human and vulnerable. Do you have any advice for aspiring girls who want to join the field? My advice for girls starting out in HR Tech is to be tenacious in achieving results, and you'll be glad to hear that there are so many opportunities for training and projects now, including through tech communities where you can find first projects and get hands-on experience.It's also great to find a mentor who has already travelled this path and will help you overcome any obstacles and move faster in your professional development. Be open, ask for advice and share your experiences – you'll be amazed at the opportunities out there! HackerNoon editorial team has launched this interview series with women in tech to celebrate their achievements and share their struggles. We need more women in technology, and by sharing stories, we can encourage many girls to follow their dreams. Share your story today! HackerNoon editorial team has launched this interview series with women in tech to celebrate their achievements and share their struggles. We need more women in technology, and by sharing stories, we can encourage many girls to follow their dreams. Share your story today! HackerNoon editorial team has launched this interview series with women in tech to celebrate their achievements and share their struggles. We need more women in technology, and by sharing stories, we can encourage many girls to follow their dreams. Share your story today! HackerNoon editorial team has launched this interview series with women in tech to celebrate their achievements and share their struggles. We need more women in technology, and by sharing stories, we can encourage many girls to follow their dreams. Share your story today ! this interview series with women in tech Share your story today Tell us about yourself! Tell us about yourself! Hi! I'm Lutsia, and I work in HR in IT. I've been doing this for over 12 years now, and I've worked with some amazing companies like gamedev, dating, music-tech, and social. Now, I'm the HR director of an international company that's doing well developing cool mobile apps for social media. My job is all about making sure the company has a great HR strategy, taking care of all the employees, and helping the business grow by developing its people. My company's products are doing well (currently, only Tinder has more installs than us) and we're seeing over 6 million installs every month. I am an active mentor in various communities such as Women in Tech, Miranna.io, ADP.list, I have done more than 100 consultations and helped more than 100 people to achieve their goals in IT. Why did you choose this field in the first place? Why did you choose this field in the first place? I've always loved studying people and watching talent grow, so it's a perfect fit for me. In HR, it's all about communication, and in the tech industry, it's a bit of a challenge because there are so many introverts who need individual attention. I've got a Master's in Human Resources Management, and I absolutely love what I do. What tech are you most worried about right now and why? What tech are you most worried about right now and why? I'm really interested in technologies that can predict how long employees will stay with a company (like prediction models for employee tenure) and assessing soft skills. Human psychology is such a complex system, isn't it? I find the standard approaches don't always work, which makes it a constant challenge in the world of people management to assess team effectiveness. What are your hobbies and interests outside of tech? What are your hobbies and interests outside of tech? When I'm not doing that, I love nothing more than spending time with my family and friends, or enjoying a good book. I'm a music lover – playing the keyboard and singing are two of my favourite pastimes. I'm also a keen outdoorsy person, and I love nothing more than a good hike! A couple of years ago, I discovered the wonderful world of mentoring, which is a really cool side activity that energises me by interacting with talented people and being useful. Let's talk about breaking the glass ceiling. What were the biggest challenges you faced as a woman in tech, and how did you deal with them? Let's talk about breaking the glass ceiling. What were the biggest challenges you faced as a woman in tech, and how did you deal with them? I started my career in the tech industry, in startups and scale-ups, so I had direct contact with founders and was lucky to avoid unpleasant situations with the glass ceiling. But I would like to see more girls among CEOs and founders of companies. I plan to launch my own project one day. Any questionable misogynistic story/situation you faced/handled, and you want to share with the HackerNoon Fam? Any questionable misogynistic story/situation you faced/handled, and you want to share with the HackerNoon Fam? When I was just starting out in my career, over 12 years ago now, I was working in talent acquisition. I remember how hiring managers would sometimes reject female candidates for developer positions just because they were girls, or because they thought programming was just not a job for women. I remember my first driving instructor always telling me that driving was not for women, but I don't encounter this now. As a HR leader, I try to nip such stories in the bud. What was the biggest setback/failure that you faced, and how did you manage it? What was the biggest setback/failure that you faced, and how did you manage it? My biggest failure was my personal and professional burnout in 2019. I was deeply immersed in the topic and clearly knew the markers of this condition, but unfortunately, I was not able to prevent it in time. However, I learnt my lesson, and I am now much more attentive to my energy levels and have improved my delegating skills even more! What's your biggest achievement that you're really proud of? What's your biggest achievement that you're really proud of? I have reached a high position just in 7 years - HR director in a fast-growing global technology company, every day I do what I love and develop not only the team, overseeing the unlocking of immense human potential, but also being involved in creating products that make a difference in this world and make life better for millions of people. I have extensive mentoring experience and have helped hundreds of people find their successful path to self-realisation in the industry. I am surrounded by amazingly smart and caring people who create beautiful things. In your opinion, why do we see this huge gender gap in the tech industry, and how can we reduce it? In your opinion, why do we see this huge gender gap in the tech industry, and how can we reduce it? The tech industry is still pretty new, so I think there are some historical and cultural things going on that are holding it back. But I do see a lot of tech companies trying to make things better, which is great. They're putting in place these DEI policies, which is really important. I'm part of a group that supports women in IT, and I think it's a vital mission. I would love to see more support, grants, quotas and investments for start-ups created by women.It would be great to change the media narrative and promote the idea that technology is a universal field where gender doesn't matter. We should be creating content that's focused on engaging women and breaking stereotypes. From a company perspective and as a C-Level, it's important to implement training to overcome unconscious bias. Who is your tech idol? Why? Who is your tech idol? Why? I really look up to Sheryl Sandberg, ex-COO of Meta and author of Lean In: Women, Work, and the Will to Lead. She started the global Lean In movement, which connects women through professional networks, support circles, and educational programs. This movement helps women in IT find like-minded people, develop skills and build careers. I really admire how she's so open about the personal struggles she's had, including the loss of her husband. It's so inspiring to see how she's managed to be so successful and still remain human and vulnerable. Do you have any advice for aspiring girls who want to join the field? Do you have any advice for aspiring girls who want to join the field? My advice for girls starting out in HR Tech is to be tenacious in achieving results, and you'll be glad to hear that there are so many opportunities for training and projects now, including through tech communities where you can find first projects and get hands-on experience.It's also great to find a mentor who has already travelled this path and will help you overcome any obstacles and move faster in your professional development. Be open, ask for advice and share your experiences – you'll be amazed at the opportunities out there!

HR Director Lutsia Gorelova Reflects on What's Changed In IT Hiring Over The Past Few Decades

The tech world relies heavily on constructive criticism to evolve and make valuable changes. Innovation requires feedback from different perspectives that can assist in creating unique solutions. Hackathons and academic conferences are places where tech leaders can put their products to the test, showcasing their utility and desire to improve solutions. Experienced tech professionals like Avinab (Avi) Singh contribute to the space by guiding the next generation of designers and technologists. His participation as an academic journal reviewer and hackathon judge contributes to the continuous evolution of tech. As the use of AI rises in design and technology fields, creators require experienced professionals who can provide insight into both user needs and advanced technological capabilities. Early Career Challenges Avi began his journey in the tech industry in Chennai, India. He graduated with a Bachelor’s degree in Telecommunication Engineering but Avi wanted cutting edge education. He decided to move to the United States, where he pursued a master’s degree in Industrial Engineering, focusing on human factors at North Carolina State University in Raleigh, North Carolina. After graduate school, Avi moved to Silicon Valley with hopes of growing his professional career. However, he was left discouraged by the fact he didn’t have any mentors to guide him along the way. Undaunted, Avi persisted in his career to great success. It was after this long journey that Avi realized he wanted to provide the mentorship he never had to young professionals who were struggling. He doesn’t want anyone to feel left behind in the already competitive world of tech. Mentoring Up-and-Coming Designers Avi’s role as a mentor within his companies and through ADPList.org helps others aspire to make significant contributions in the industry. He has accumulated over 800 minutes of mentoring across 3 countries, demonstrating his desire to help professionals grow. His work in assisting global hiring initiatives through internship programs at EY GigNow and Dusty Robotics has jumpstarted the careers of novice designers and helped them navigate ongoing professional challenges & grow their careers. Providing mentorship to young professionals gives them the tools and insights they need to succeed in the tech industry. Having a competitive advantage against others in the industry is vital to a successful career. As a result of these internship programs, many junior designers have successfully secured full-time roles in these companies, contributing to the teams’ growth and innovation with the addition of their fresh perspectives. Serving as a Judge at Hackathons Avi’s role as a judge at notable hackathon events like HackMIT, SwiftHack, and GenLab2 has played a crucial role in bringing novel ideas to the tech industry. His experience allows him to provide insightful advice for novices and solidifies his longstanding knowledge in the field. At hackathons, Avi is responsible for evaluating new projects and guiding teams toward creating practical and impactful tech solutions. His judging experience helps him identify promising talent and encourages collaboration among participants. Reviewer for Reputed Journals and Conferences Avi is also an active reviewer for top-tier conferences such as ACM and IEEE in the field of Human-Computer Interaction and AI/ML. This growing industry is in need of the insight of experienced professionals to ground itself in tech. His ongoing involvement in reviewing scientific papers for major conferences has gained him recognition as a trusted reviewer by UIST and AutomotiveUI. He continues to review papers at other notable conferences like CHI, HAI, ICSI, NordiCHI, and UbiComp/ISWC, among others. Peer review is essential to maintaining the quality of research needed to develop impactful solutions and promote technological advancements. Avi’s contributions help ensure that emerging research aligns with high industry standards by ensuring paper reviews shape the current challenges in design and technology. How Avi Singh is Shaping Future Tech Leaders Through the mentorship of novice designers, reviewing academic papers, and judging at hackathons, Avi is actively contributing to the development of future leaders and their solutions in the tech industry. Avi is currently working as a Staff Product Designer at Moveworks, where he works on Generative AI tools to help small to large enterprises scale their employee experience organizations. His latest creation is a Generative AI tool called Brief Me that assists users with summarizing and question & answer with dense documents. Brief Me helps increase employee productivity by helping them read large documents, create reports and write emails using content from these documents. Avi creates software used daily by employees of Fortune 100 companies. Avi continues to provide valuable solutions to the field of Design of AI/ML products through his work and mentorship initiatives. Sharing knowledge and experience is vital for overcoming the challenges newcomers face in the field. Up-and-coming talents are more likely to succeed in the industry with the help of mentors like Avi. Avi’s efforts to create a collaborative, supportive environment for emerging professionals in both corporate and educational settings. His dedication to mentorship aims to alleviate the difficulties he faced early in his career by providing others with the guidance they need to thrive. The tech world relies heavily on constructive criticism to evolve and make valuable changes. Innovation requires feedback from different perspectives that can assist in creating unique solutions. Hackathons and academic conferences are places where tech leaders can put their products to the test, showcasing their utility and desire to improve solutions. Experienced tech professionals like Avinab (Avi) Singh contribute to the space by guiding the next generation of designers and technologists. His participation as an academic journal reviewer and hackathon judge contributes to the continuous evolution of tech. As the use of AI rises in design and technology fields, creators require experienced professionals who can provide insight into both user needs and advanced technological capabilities. Early Career Challenges Early Career Challenges Avi began his journey in the tech industry in Chennai, India. He graduated with a Bachelor’s degree in Telecommunication Engineering but Avi wanted cutting edge education. He decided to move to the United States, where he pursued a master’s degree in Industrial Engineering, focusing on human factors at North Carolina State University in Raleigh, North Carolina. After graduate school, Avi moved to Silicon Valley with hopes of growing his professional career. However, he was left discouraged by the fact he didn’t have any mentors to guide him along the way. Undaunted, Avi persisted in his career to great success. It was after this long journey that Avi realized he wanted to provide the mentorship he never had to young professionals who were struggling. He doesn’t want anyone to feel left behind in the already competitive world of tech. Mentoring Up-and-Coming Designers Mentoring Up-and-Coming Designers Avi’s role as a mentor within his companies and through ADPList.org helps others aspire to make significant contributions in the industry. He has accumulated over 800 minutes of mentoring across 3 countries, demonstrating his desire to help professionals grow. ADPList.org ADPList.org His work in assisting global hiring initiatives through internship programs at EY GigNow and Dusty Robotics has jumpstarted the careers of novice designers and helped them navigate ongoing professional challenges & grow their careers. EY GigNow EY GigNow Dusty Robotics Dusty Robotics Providing mentorship to young professionals gives them the tools and insights they need to succeed in the tech industry. Having a competitive advantage against others in the industry is vital to a successful career. As a result of these internship programs, many junior designers have successfully secured full-time roles in these companies, contributing to the teams’ growth and innovation with the addition of their fresh perspectives. Serving as a Judge at Hackathons Serving as a Judge at Hackathons Avi’s role as a judge at notable hackathon events like HackMIT, SwiftHack, and GenLab2 has played a crucial role in bringing novel ideas to the tech industry. His experience allows him to provide insightful advice for novices and solidifies his longstanding knowledge in the field. At hackathons, Avi is responsible for evaluating new projects and guiding teams toward creating practical and impactful tech solutions. His judging experience helps him identify promising talent and encourages collaboration among participants. Reviewer for Reputed Journals and Conferences Reviewer for Reputed Journals and Conferences Avi is also an active reviewer for top-tier conferences such as ACM and IEEE in the field of Human-Computer Interaction and AI/ML. This growing industry is in need of the insight of experienced professionals to ground itself in tech. His ongoing involvement in reviewing scientific papers for major conferences has gained him recognition as a trusted reviewer by UIST and AutomotiveUI. He continues to review papers at other notable conferences like CHI, HAI, ICSI, NordiCHI, and UbiComp/ISWC, among others. Peer review is essential to maintaining the quality of research needed to develop impactful solutions and promote technological advancements. Avi’s contributions help ensure that emerging research aligns with high industry standards by ensuring paper reviews shape the current challenges in design and technology. How Avi Singh is Shaping Future Tech Leaders How Avi Singh is Shaping Future Tech Leaders Through the mentorship of novice designers, reviewing academic papers, and judging at hackathons, Avi is actively contributing to the development of future leaders and their solutions in the tech industry. Avi is currently working as a Staff Product Designer at Moveworks , where he works on Generative AI tools to help small to large enterprises scale their employee experience organizations. His latest creation is a Generative AI tool called Brief Me that assists users with summarizing and question & answer with dense documents. Brief Me helps increase employee productivity by helping them read large documents, create reports and write emails using content from these documents. Avi creates software used daily by employees of Fortune 100 companies. Avi continues to provide valuable solutions to the field of Design of AI/ML products through his work and mentorship initiatives. Moveworks Moveworks Sharing knowledge and experience is vital for overcoming the challenges newcomers face in the field. Up-and-coming talents are more likely to succeed in the industry with the help of mentors like Avi. Avi’s efforts to create a collaborative, supportive environment for emerging professionals in both corporate and educational settings. His dedication to mentorship aims to alleviate the difficulties he faced early in his career by providing others with the guidance they need to thrive.

Shaping Future Tech Leaders: Avinab Singh’s Role as a Mentor and Reviewer

Jutïr Tech irpirinakar uñstayaña: Avinab Singh jupan lurawipax mä Mentor ukhamarak Revisor ukhamawa

Forjando a los futuros líderes tecnológicos: el papel de Avinab Singh como mentor y revisor

Ապագա տեխնոլոգիական առաջնորդների ձևավորում. Ավինաբ Սինգհի դերը որպես դաստիարակ և գրախոս

未来のテクノロジーリーダーを育てる: メンターおよびレビュアーとしてのアヴィナブ・シンの役割

მომავლის ტექნიკური ლიდერების ჩამოყალიბება: ავინაბ სინგჰის როლი, როგორც მენტორი და მიმომხილველი

미래 기술 리더 형성: 멘토 및 리뷰어로서의 Avinab Singh의 역할

Būsimų technologijų lyderių formavimas: Avinabo Singho, kaip mentoriaus ir apžvalgininko, vaidmuo

Обликување идни технолошки лидери: Улогата на Авинаб Синг како ментор и рецензент

Membentuk Pemimpin Teknologi Masa Depan: Peranan Avinab Singh sebagai Mentor dan Pengulas

Kształtowanie przyszłych liderów technologicznych: rola Avinaba Singha jako mentora i recenzenta

Ташаккули пешвоёни ояндаи техникӣ: Нақши Авинаб Сингҳ ҳамчун мураббӣ ва баррасӣ

การสร้างผู้นำด้านเทคโนโลยีในอนาคต: บทบาทของ Avinab Singh ในฐานะที่ปรึกษาและผู้ตรวจสอบ

Changing the Face of Web Management: Interview with Startups of the Year 2024 Nominee, SiteSkite

This paper is under CC 4.0 license. available on arxiv Authors: (1) Joel Jang, CarperAI,University of Washington & Allen Institute for AI; (2) Seungone Kim, KAIST AI; (3) Yizhong Wang, University of Washington; (4) Jack Hessel, University of Washington; (5) Luke Zettlemoyer, Aleph Alpha; (6) Hannaneh Hajishirzi, University of Washington & Allen Institute for AI; (7) Yejin Choi, UC San Diego. 5 CONCLUSION Previous work has shown that adapting LLMs with RLHF helps them generate outputs that are preferred by humans over the supervised fine-tuned counterpart. However, recent work has also pointed out that simply training LLMs to abide by the preference of the general may result in ignoring individual preferences and values. In this work, we provide the first steps to tackle this issue by proposing Reinforcement Learning from Personalized Human Feedback as a multi-objective problem so that LLMs can be aligned to follow conflicting preferences. We propose a promising method called P-SOUPS that is able to composite models trained on single objectives on the fly during inference. We also highlight the scalability of P-SOUPS by showing that it scales linearly, instead of exponentially like the MORL baseline, with regards to the number of new preferences, which is required to provide true personalization to individual users. ACKNOWLEDGMENTS Thanks to Minyoung Hwang, Sungdong Kim, Tim Dettmers, Yoonjoo Lee, and Margaret Li for helpful feedback and discussion. REFERENCES Xiang Ao, Xiting Wang, Ling Luo, Ying Qiao, Qing He, and Xing Xie. PENS: A dataset and generic framework for personalized news headline generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pp. 82–92, Online, August 2021. Association for Computational Linguistics. doi: 10.18653/v1/2021.acl-long.7. URL https: //aclanthology.org/2021.acl-long.7. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, T. J. Henighan, Nicholas Joseph, Saurav Kadavath, John Kernion, Tom Conerly, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Tristan Hume, Scott Johnston, Shauna Kravec, Liane Lovitt, Neel Nanda, Catherine Olsson, Dario Amodei, Tom B. Brown, Jack Clark, Sam McCandlish, Christopher Olah, Benjamin Mann, and Jared Kaplan. Training a helpful and harmless assistant with reinforcement learning from human feedback. ArXiv, abs/2204.05862, 2022a. URL https://api.semanticscholar.org/ CorpusID:248118878. Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, et al. Constitutional ai: Harmlessness from ai feedback. arXiv preprint arXiv:2212.08073, 2022b. Stephen Casper, Xander Davies, Claudia Shi, Thomas Krendl Gilbert, Jer´ emy Scheurer, Javier ´ Rando, Rachel Freedman, Tomasz Korbak, David Lindner, Pedro Freire, et al. Open problems and fundamental limitations of reinforcement learning from human feedback. arXiv preprint arXiv:2307.15217, 2023. Deborah Cohen, Moonkyung Ryu, Yinlam Chow, Orgad Keller, Ido Greenberg, Avinatan Hassidim, Michael Fink, Yossi Matias, Idan Szpektor, Craig Boutilier, et al. Dynamic planning in openended dialogue using reinforcement learning. arXiv preprint arXiv:2208.02294, 2022. Shachar Don-Yehiya, Elad Venezian, Colin Raffel, Noam Slonim, Yoav Katz, and Leshem Choshen. Cold fusion: Collaborative descent for distributed multitask finetuning. arXiv preprint arXiv:2212.01378, 2022. Yann Dubois, Xuechen Li, Rohan Taori, Tianyi Zhang, Ishaan Gulrajani, Jimmy Ba, Carlos Guestrin, Percy Liang, and Tatsunori B Hashimoto. Alpacafarm: A simulation framework for methods that learn from human feedback. arXiv preprint arXiv:2305.14387, 2023. Shangbin Feng, Chan Young Park, Yuhan Liu, and Yulia Tsvetkov. From pretraining data to language models to downstream tasks: Tracking the trails of political biases leading to unfair nlp models. arXiv preprint arXiv:2305.08283, 2023. Xinyang Geng, Arnav Gudibande, Hao Liu, Eric Wallace, Pieter Abbeel, Sergey Levine, and Dawn Song. Koala: A dialogue model for academic research. Blog post, April, 1, 2023. Conor F Hayes, Roxana Radulescu, Eugenio Bargiacchi, Johan K ˘ allstr ¨ om, Matthew Macfarlane, ¨ Mathieu Reymond, Timothy Verstraeten, Luisa M Zintgraf, Richard Dazeley, Fredrik Heintz, et al. A practical guide to multi-objective reinforcement learning and planning. Autonomous Agents and Multi-Agent Systems, 36(1):26, 2022. Edward J Hu, yelong shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. LoRA: Low-rank adaptation of large language models. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum? id=nZeVKeeFYf9. Chengsong Huang, Qian Liu, Bill Yuchen Lin, Tianyu Pang, Chao Du, and Min Lin. Lorahub: Efficient cross-task generalization via dynamic lora composition. ArXiv, abs/2307.13269, 2023. URL https://api.semanticscholar.org/CorpusID:260155012. Gabriel Ilharco, Marco Tulio Ribeiro, Mitchell Wortsman, Suchin Gururangan, Ludwig Schmidt, Hannaneh Hajishirzi, and Ali Farhadi. Editing models with task arithmetic. ICLR, 2022. Joel Jang, Seungone Kim, Seonghyeon Ye, Doyoung Kim, Lajanugen Logeswaran, Moontae Lee, Kyungjae Lee, and Minjoon Seo. Exploring the benefits of training expert language models over instruction tuning. In Andreas Krause, Emma Brunskill, Kyunghyun Cho, Barbara Engelhardt, Sivan Sabato, and Jonathan Scarlett (eds.), Proceedings of the 40th International Conference on Machine Learning, volume 202 of Proceedings of Machine Learning Research, pp. 14702–14729. PMLR, 23–29 Jul 2023. URL https://proceedings.mlr.press/v202/jang23a. html. Sungdong Kim, Sanghwan Bae, Jamin Shin, Soyoung Kang, Donghyun Kwak, Kang Min Yoo, and Minjoon Seo. Aligning large language models through synthetic feedback. arXiv preprint arXiv:2305.13735, 2023. Hannah Rose Kirk, Bertie Vidgen, Paul Rottger, and Scott A Hale. Personalisation within bounds: A ¨ risk taxonomy and policy framework for the alignment of large language models with personalised feedback. arXiv preprint arXiv:2303.05453, 2023. Noriyuki Kojima, Alane Suhr, and Yoav Artzi. Continual learning for grounded instruction generation by observing human following behavior. Transactions of the Association for Computational Linguistics, 9:1303–1319, 2021. Margaret Li, Suchin Gururangan, Tim Dettmers, Mike Lewis, Tim Althoff, Noah A Smith, and Luke Zettlemoyer. Branch-train-merge: Embarrassingly parallel training of expert language models. arXiv preprint arXiv:2208.03306, 2022. Pan Li and Alexander Tuzhilin. Towards controllable and personalized review generation. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 3237–3245, Hong Kong, China, November 2019. Association for Computational Linguistics. doi: 10.18653/v1/D19-1319. URL https://aclanthology.org/D19-1319. Bodhisattwa Prasad Majumder, Shuyang Li, Jianmo Ni, and Julian McAuley. Generating personalized recipes from historical user preferences. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pp. 5976–5982, Hong Kong, China, November 2019. Association for Computational Linguistics. doi: 10.18653/v1/D19-1613. URL https://aclanthology.org/D19-1613. Pierre-Emmanuel Mazare, Samuel Humeau, Martin Raison, and Antoine Bordes. Training mil- ´ lions of personalized dialogue agents. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pp. 2775–2779, Brussels, Belgium, OctoberNovember 2018. Association for Computational Linguistics. doi: 10.18653/v1/D18-1298. URL https://aclanthology.org/D18-1298. Jacob Menick, Maja Trebacz, Vladimir Mikulik, John Aslanides, Francis Song, Martin Chadwick, Mia Glaese, Susannah Young, Lucy Campbell-Gillingham, Geoffrey Irving, et al. Teaching language models to support answers with verified quotes. arXiv preprint arXiv:2203.11147, 2022. Reiichiro Nakano, Jacob Hilton, S. Arun Balaji, Jeff Wu, Ouyang Long, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, Xu Jiang, Karl Cobbe, Tyna Eloundou, Gretchen Krueger, Kevin Button, Matthew Knight, Benjamin Chess, and John Schulman. Webgpt: Browser-assisted question-answering with human feedback. ArXiv, abs/2112.09332, 2021a. URL https://api.semanticscholar.org/CorpusID:245329531. Reiichiro Nakano, Jacob Hilton, Suchir Balaji, Jeff Wu, Long Ouyang, Christina Kim, Christopher Hesse, Shantanu Jain, Vineet Kosaraju, William Saunders, et al. Webgpt: Browser-assisted question-answering with human feedback. arXiv preprint arXiv:2112.09332, 2021b. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke E. Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Francis Christiano, Jan Leike, and Ryan J. Lowe. Training language models to follow instructions with human feedback. ArXiv, abs/2203.02155, 2022a. URL https://api.semanticscholar.org/ CorpusID:246426909. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35: 27730–27744, 2022b. Baolin Peng, Chunyuan Li, Pengcheng He, Michel Galley, and Jianfeng Gao. Instruction tuning with gpt-4. arXiv preprint arXiv:2304.03277, 2023. Jonas Pfeiffer, Naman Goyal, Xi Victoria Lin, Xian Li, James Cross, Sebastian Riedel, and Mikel Artetxe. Lifting the curse of multilinguality by pre-training modular transformers. arXiv preprint arXiv:2205.06266, 2022. Alexandre Rame, Guillaume Couairon, Mustafa Shukor, Corentin Dancette, Jean-Baptiste Gaya, Laure Soulier, and Matthieu Cord. Rewarded soups: towards pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards. arXiv preprint arXiv:2306.04488, 2023. Alireza Salemi, Sheshera Mysore, Michael Bendersky, and Hamed Zamani. Lamp: When large language models meet personalization. arXiv preprint arXiv:2304.11406, 2023. Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee, Percy Liang, and Tatsunori Hashimoto. Whose opinions do language models reflect? arXiv preprint arXiv:2303.17548, 2023. John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347, 2017. Umer Siddique, Paul Weng, and Matthieu Zimmer. Learning fair policies in multi-objective (deep) reinforcement learning with average and discounted rewards. In International Conference on Machine Learning, pp. 8905–8915. PMLR, 2020. David Silver, Satinder Singh, Doina Precup, and Richard S Sutton. Reward is enough. Artificial Intelligence, 299:103535, 2021. Prasann Singhal, Tanya Goyal, Jiacheng Xu, and Greg Durrett. A long way to go: Investigating length correlations in rlhf. arXiv preprint arXiv:2310.03716, 2023. Feifan Song, Bowen Yu, Minghao Li, Haiyang Yu, Fei Huang, Yongbin Li, and Houfeng Wang. Preference ranking optimization for human alignment. arXiv preprint arXiv:2306.17492, 2023. Nisan Stiennon, Long Ouyang, Jeffrey Wu, Daniel Ziegler, Ryan Lowe, Chelsea Voss, Alec Radford, Dario Amodei, and Paul F Christiano. Learning to summarize with human feedback. Advances in Neural Information Processing Systems, 33:3008–3021, 2020. Alane Suhr and Yoav Artzi. Continual learning for instruction following from realtime feedback. arXiv preprint arXiv:2212.09710, 2022. Richard S Sutton and Andrew G Barto. Reinforcement learning: An introduction. MIT press, 2018. Romal Thoppilan, Daniel De Freitas, Jamie Hall, Noam Shazeer, Apoorv Kulshreshtha, Heng-Tze Cheng, Alicia Jin, Taylor Bos, Leslie Baker, Yu Du, et al. Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239, 2022. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothee´ Lacroix, Baptiste Roziere, Naman Goyal, Eric Hambro, Faisal Azhar, et al. Llama: Open and ` efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023. Peter Vamplew, Richard Dazeley, Cameron Foale, Sally Firmin, and Jane Mummery. Humanaligned artificial intelligence is a multiobjective problem. Ethics and Information Technology, 20:27–40, 2018. Kristof Van Moffaert and Ann Nowe. Multi-objective reinforcement learning using sets of pareto ´ dominating policies. The Journal of Machine Learning Research, 15(1):3483–3512, 2014. Kristof Van Moffaert, Madalina M Drugan, and Ann Nowe. Scalarized multi-objective reinforce- ´ ment learning: Novel design techniques. In 2013 IEEE symposium on adaptive dynamic programming and reinforcement learning (ADPRL), pp. 191–199. IEEE, 2013. Yizhong Wang, Hamish Ivison, Pradeep Dasigi, Jack Hessel, Tushar Khot, Khyathi Raghavi Chandu, David Wadden, Kelsey MacMillan, Noah A Smith, Iz Beltagy, et al. How far can camels go? exploring the state of instruction tuning on open resources. arXiv preprint arXiv:2306.04751, 2023. Mitchell Wortsman, Suchin Gururangan, Shen Li, Ali Farhadi, Ludwig Schmidt, Michael Rabbat, and Ari S Morcos. lo-fi: distributed fine-tuning without communication. arXiv preprint arXiv:2210.11948, 2022a. Mitchell Wortsman, Gabriel Ilharco, Samir Ya Gadre, Rebecca Roelofs, Raphael Gontijo-Lopes, Ari S Morcos, Hongseok Namkoong, Ali Farhadi, Yair Carmon, Simon Kornblith, et al. Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time. In International Conference on Machine Learning, pp. 23965–23998. PMLR, 2022b. Jeff Wu, Long Ouyang, Daniel M Ziegler, Nisan Stiennon, Ryan Lowe, Jan Leike, and Paul Christiano. Recursively summarizing books with human feedback. arXiv preprint arXiv:2109.10862, 2021a. Yuwei Wu, Xuezhe Ma, and Diyi Yang. Personalized response generation via generative split memory network. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pp. 1956–1970, 2021b. Zeqiu Wu, Yushi Hu, Weijia Shi, Nouha Dziri, Alane Suhr, Prithviraj Ammanabrolu, Noah A Smith, Mari Ostendorf, and Hannaneh Hajishirzi. Fine-grained human feedback gives better rewards for language model training. arXiv preprint arXiv:2306.01693, 2023. Jie Xu, Yunsheng Tian, Pingchuan Ma, Daniela Rus, Shinjiro Sueda, and Wojciech Matusik. Prediction-guided multi-objective reinforcement learning for continuous robot control. In International conference on machine learning, pp. 10607–10616. PMLR, 2020. Jing Xu, Arthur Szlam, and Jason Weston. Beyond goldfish memory: Long-term open-domain conversation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 5180–5197, Dublin, Ireland, May 2022. Association for Computational Linguistics. doi: 10.18653/v1/2022.acl-long.356. URL https: //aclanthology.org/2022.acl-long.356. Runzhe Yang, Xingyuan Sun, and Karthik Narasimhan. A generalized algorithm for multi-objective reinforcement learning and policy adaptation. Advances in neural information processing systems, 32, 2019. Saizheng Zhang, Emily Dinan, Jack Urbanek, Arthur Szlam, Douwe Kiela, and Jason Weston. Personalizing dialogue agents: I have a dog, do you have pets too? arXiv preprint arXiv:1801.07243, 2018. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric Xing, et al. Judging llm-as-a-judge with mt-bench and chatbot arena. arXiv preprint arXiv:2306.05685, 2023. Yinhe Zheng, Guanyi Chen, Minlie Huang, Song Liu, and Xuan Zhu. Personalized dialogue generation with diversified traits. arXiv preprint arXiv:1901.09672, 2019. Daniel M Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B Brown, Alec Radford, Dario Amodei, Paul Christiano, and Geoffrey Irving. Fine-tuning language models from human preferences. arXiv preprint arXiv:1909.08593, 2019. A DETAILS OF EVALUATION SETUP We use a modified version of the GPT4 annotation prompt used by Dubois et al. (2023). We modify the criteria to perform the pairwise evaluation from general to a single preference dimension. We also provide 3 demonstrations: one scenario where there is a tie because both responses do not contain any notion of the preference (e.g. both responses do not show any signs of friendliness), one scenario where there is a clear winner, and one scenario where they are both good, but one is better than the other. Simulated pairwise evaluation We recruited 24 crowd workers for our human evaluation. Figure 5 shows the interface used for human evaluation. We consider both the ‘Tie’ and ‘Both are bad’ options to be Ties. B CRITERIA-WISE EVALUATION The criteria-wise win rate (%) across all the methods are shown in Figure 6. The criteria-wise win rate was calculated by getting the average win-rate of the preference combinations that contained the specific preference dimension. For example, when calculating the criteria-wise win rate of ‘Elementary’, we got the average win-rate of the preference combinations that contained the ‘Elementary’ preference, which includes AAA, AAB, ABA, and ABB. C THE FULL LIST OF EVALUATION PROMPTS The full list of evaluation prompts used in our experiments are provided in Table 5 D DETAILED RESULTS FOR HUMAN EVALUATION AND GPT-4 EVALUATION We provide detailed results (win / loss / tie for each of the preference combinations for our main experimental results. Table 6 shows the GPT4 evaluation and Table 7 shows the human evaluation results. E EXAMPLES OF P-SOUPS TEXT GENERATIONS Table 8 shows empirical examples of the text generated from each preference combination of the 16 preference combination experiments for the same prompt.

Personalized Soups: LLM Alignment Via Parameter Merging - Conclusion & References

Imposter Syndrome Is an Invention: Interview With Angelina Severino, Design Lead at Gamgee Amsterdam

Stories, News & Mentions about ADP

Categories

Trending Topics

Stories, News & Mentions about ADP

Adp’s Stories on HackerNoon

Adp's News & Mentions

Categories

Trending Topics