Gagan Bansal @bansalg_ Twitter profile

Pinned Tweet

Gagan Bansal

7 months

Working on or interested on agents based on Large Language Models like GPT-4? Follow @pyautogen to get latest news and use cases of our rapidly expanding multi-agent framework #AutoGen !

0

12

Last Seen Profiles

@nf_suzanne

@KristelAlver

@MECAForPeace

@limnios1234

@RMUWBasketball

@Forgot3thWorlds

@sheffieldtrib

@BRSparty

@wallacemenace

@PTeamfinder

@StCenyddEnglish

@ima_scho

@ladaamores

@Donnami_cos

@pinkolate_

@DavidBlackShoes

@filip_dragovic

@GeeceeKoncept

@DanielA54597987

@Ot1297

@Gerick_6

@jbradfordmusic

@olivier_pourret

@FlukeTheWolf

@CleeTompkinson

@jay_miller66

@CollegeFBHoF

@KiwiFarmsLs

@ggrudic

@navithadev

@StubbyPrice

@SFUnified

@CoachChuy67

@PartnersUcu

@audrey_crespo

@w1lko

Gagan Bansal

@bansalg_

2 years

Looking for research internships on #AI + #HCI ?? Our team, Human-AI eXperiences (HAX) at @MSFTResearch , is hiring multiple interns for summer 2023 on topics related to human-centered AI (see 👇). cc: @SaleemaAmershi @vykthur @adamfourney Apply at

10

72

247

Gagan Bansal

@bansalg_

4 years

New work on explainable AI! w/ @tongshuangwu , J. Zhu, R. Fok, @besanushi , @ecekamar , M. Ribeiro, and @dsweld . When AIs advise people, does an explanation of its reasoning actually help the person? Does it let the human outperform the AI? Does it ...(1/6)

Does the Whole Exceed its Parts? The Effect of AI Explanations on...

Many researchers motivate explainable AI with studies showing that human-AI team performance on decision-making tasks improves when the AI explains its recommendations. However, prior studies...

arxiv.org

6

43

217

Gagan Bansal

@bansalg_

6 months

📢📢We are also hiring summer interns!📢📢 Come work with our teams in MSR AI Frontiers to conduct research on interaction between AI agents and people!

8

39

214

Gagan Bansal

@bansalg_

5 months

Our team will start reviewing internship applications on Jan 5th 2024! So please apply and submit material before that!

Gagan Bansal

@bansalg_

6 months

📢📢We are also hiring summer interns!📢📢 Come work with our teams in MSR AI Frontiers to conduct research on interaction between AI agents and people!

8

39

214

2

25

154

Gagan Bansal

@bansalg_

2 years

🚨Announcing #CHI2022 workshop on Trust and Reliance in AI-Human Teams (TRAIT) ft. keynote by @jdlee888 , panel w/ #HCI & #AI experts, group activities & more! w/ @alison_m_smith , @ZanaBucinca , @tongshuangwu , @d19fe8 , @JessicaHullman , @DrSimoneStumpf See:

3

30

123

Gagan Bansal

@bansalg_

2 years

About 60 researchers brainstorming how to define, measure, and shape trust and reliance in human-AI interaction at the #TRAIT2022 hybrid workshop at #CHI2022 @sigchi

3

21

106

Gagan Bansal

@bansalg_

1 year

🚨 Working on #AI + #HCI ? Join the 2nd #CHI2023 workshop on Trust & Reliance in AI-Human Teams (TRAIT), focused on real-world human-AI use cases. Keynote: @Carryveggies & Michael Terry. w/ @alison_m_smith @ZanaBucinca @tongshuangwu @d19fe8 @JessicaHullman @DrSimoneStumpf

1

29

96

Gagan Bansal

@bansalg_

4 years

Excited to share a draft of our new work on human-centered AI! w/ @besanushi @ecekamar @erichorvitz @dsweld When an AI assists human decision-makers, e.g, by recommending its predictions, is the most accurate AI necessarily the best team-mate? (1/5)

2

18

84

Gagan Bansal

@bansalg_

2 years

And its a wrap! We had 80 ppl/8 timezones join in a hybrid setting. Thanks to my co-organizers, speakers, pc, & participants for making this workshop possible @alison_m_smith @ZanaBucinca @tongshuangwu @d19fe8 @JessicaHullman @DrSimoneStumpf #TRAIT2022 #CHI2022

0

11

77

Gagan Bansal

@bansalg_

2 years

Excited to be at #CHI2022 in-person!! DM me if you'd like to catch up there! We can talk about human-AI interaction (or not)!

2

0

68

Gagan Bansal

@bansalg_

6 months

Agents w/ APIs/tools are cool and popular now but throwback to one of my favorite classic papers from 1994 on agents by @etzioni and @dsweld The first time I read and obsessed about it was when I was an undergraduate and saw it as a reference in Russel and Norvig!

4

10

52

Gagan Bansal

@bansalg_

3 years

Do you work on AI/ML + HCI? We invite submissions for our new journal--- Special Issue on "AI for (and by) the People" Wide range of human + AI topics ✅ Open-source ✅ Virtual workshop post publication ✅ cc: @alison_m_smith , @gonzaloworks

Alison Renner (Smith)

@alison_m_smith

3 years

Working on interesting problems around humans & #ai ? The "AI for (and by) the People" journal special issue explores the opportunities and challenges of designing and developing AI/ML for people. Deadline: Sep 15. #hcai #hcml #cfp @bansalg_ @gonzaloworks

0

3

10

2

23

43

Gagan Bansal

@bansalg_

7 months

I am super excited to see how this will power more research on not just multi agents workflows but entire new subtopics in human-AI interaction! Checkout our new paper below ⬇️

Chi Wang

@Chi_Wang_

7 months

Imagine if ✨multiple✨ ChatGPT agents could collaborate to solve complex tasks for you! 🧑‍🦱🤝🤖🤖🤖 📢 AutoGen: A new framework for building multi-agent LLM applications It allows creating many agents that converse to solve complex tasks! ... 1/4

7

50

190

1

10

40

Gagan Bansal

@bansalg_

2 years

My academic Twitter colleagues, I need a tiny favor! If you've ever made a paper co-authored by me a required reading for a course, can you please DM me the details (esp. the course # and year)? It would help me with my US immigration app! (Angy cat from last yr as clickbait)

5

4

38

Gagan Bansal

@bansalg_

1 year

We expect to begin evaluating applications for 2023 summer internships with our team next week. Apply by early next week for full consideration!

Gagan Bansal

@bansalg_

2 years

Looking for research internships on #AI + #HCI ?? Our team, Human-AI eXperiences (HAX) at @MSFTResearch , is hiring multiple interns for summer 2023 on topics related to human-centered AI (see 👇). cc: @SaleemaAmershi @vykthur @adamfourney Apply at

10

72

247

0

6

38

Gagan Bansal

@bansalg_

3 years

Looking for interdisciplinary internship on human-centered AI/ML/NLP?? 👇

Saleema Amershi

@SaleemaAmershi

3 years

**Multiple** internship opportunities to work with the #HAX Team @MSFTResearch in 2022! If you're interested in #ResponsibleAI , #AI #UX , and tools for creating these, apply here: Learn more about our team here:

4

42

149

1

2

31

Gagan Bansal

@bansalg_

11 months

Such wonderful reception of our work on understanding programmer-CoPilot interaction. Has implications for understanding human-LLM interaction in general. cc: @HsseinMzannar @adamfourney @erichorvitz

👩‍💻 Paige Bailey

@DynamicWebPaige

11 months

"Our studies revealed that when solving a coding task with Copilot, programmers may spend a large fraction of total session time (34.3%) on just double-checking and editing suggestions, and spend *more than half* of the task time on Copilot related activities, together indicating

34

345

2K

0

5

28

Gagan Bansal

@bansalg_

1 year

Checkout our new work on understanding and improving human-LLM interaction! #GitHubCopilot #LLMs

Hussein Mozannar

@HsseinMzannar

1 year

As Copilot becomes more popular, we need to understand how programmers interact with it. We built a model of interaction between Copilot and Programmers named 'CUPS' and predict programmer behavior in our latest paper

6

29

138

1

2

28

Gagan Bansal

@bansalg_

2 years

We received an overwhelming number of submissions on trust and reliance for human-AI interaction at the TRAIT workshop @ #CHI2022 Even if you are not attending the workshop on 30th April, checkout the accepted papers below!

Alison Renner (Smith)

@alison_m_smith

2 years

📢 #chi2022 TRAIT workshop program is up w/ keynote by @Jdlee888 , an all-star panel including @SaleemaAmershi @mariadearteaga @tmiller_unimelb @kgajos , and 41 papers (!) here: w/ @DrSimoneStumpf @JessicaHullman @ZanaBucinca @bansalg_ @d19fe8 @tongshuangwu

0

16

47

1

3

27

Gagan Bansal

@bansalg_

5 months

Come learn more about agents and AutoGen at the Microsoft booth at #NeurIPS2023 ! @Chi_Wang_ , @qingyun_wu , and I will be there on Monday 12/11, between 9 am-noon CST and 3:30-4:00 pm CST. Location: Booth 1003 - Next to entrance Hall D cc: @pyautogen , @MSFTResearch

AutoGen

@pyautogen

5 months

Working on AI Agents and attending #NeurIPS2023 ?? Don't miss in-person talk by @Chi_Wang_ on @pyautogen Details:

1

11

40

1

4

26

Gagan Bansal

@bansalg_

6 months

✨How to get multiple Open AI Assistants ( #GPTs ) *and* #AutoGen Agents all working together to solve tasks?✨ To learn, see video by AI Jason This is a good example of why supporting cross-platform agents will become increasingly important!

"Research agent 3.0 - Build a group of AI researchers" - Here is how

I built a team of AI agents via Autogen + GPTs, they generate high quality research & verify each other's workGet free credits to finetune your own LLM on Gr...

www.youtube.com

2

6

27

Gagan Bansal

@bansalg_

1 year

Checkout our new work on improving interaction with LLMs 👇👇👇

Helena Vasconcelos' Research

@HelenasResearch

1 year

New research on improving human-AI interaction! 🌟 LLMs like #CoPilot can be amazing! But they can also suggest erroneous code & verifying their suggestions takes effort. We show that communicating uncertainty reduces these costs! BUT the notion of uncertainty also matters. 1/4

5

24

129

0

7

24

Gagan Bansal

@bansalg_

3 years

New compelling evidence for developing explainable AI! Our user-studies on open-domain QA show that explanations help end-users and outperform calibrated confidence (strong, unbeaten baseline) by a significant margin! That too whilst achieving "complementary" performance!! (1/2)

1

3

24

Gagan Bansal

@bansalg_

6 months

“Orca, a 13-billion language model that demonstrated strong reasoning abilities by imitating the step-by-step reasoning traces of more capable LLMs.”

Microsoft Research

@MSFTResearch

6 months

At Microsoft, we’re expanding AI capabilities by training small language models to achieve the kind of enhanced reasoning and comprehension typically found only in much larger models.

70

243

2K

0

2

22

Gagan Bansal

@bansalg_

2 years

I 100% agree -- its important to understand when explanations help users, for which tasks, and along what metrics. E.g., it’s important to understand when explanations do and don’t lead to appropriate reliance... [1/2]

𝙷𝚒𝚖𝚊 𝙻𝚊𝚔𝚔𝚊𝚛𝚊𝚓𝚞

@hima_lakkaraju

2 years

Just because one user study showed that explanations produced by a method were not helpful for N homogenous users in a particular context, this does not imply that the method in question has no utility in any other setting. It is important to appreciate this nuance [4/N]

1

31

1

2

21

Gagan Bansal

@bansalg_

2 years

"The striking difference was that developers who used GitHub Copilot completed the task significantly faster–55% faster than the developers who didn’t use GitHub Copilot." Very promising, real-world results for human-AI interaction!

Eirini Kalliamvakou

@irina_kAl

2 years

Equal parts hard work and exciting work, I'm very glad to be sharing these results! #GitHubCopilot has had such strong impact on developers on many levels, it's a privilege to have front row seats to how we understand and measure that. More to come!

4

13

38

1

20

Gagan Bansal

@bansalg_

2 months

Seattle flu gave me insomnia so I thought I'd create an example of #AutoGen feature that I find useful for creating end applications. Here I wanted the agents to find recent GitHub issues on AutoGen's repo and then render a neat markdown table using @willmcgugan 's Rich library.

0

4

22

Gagan Bansal

@bansalg_

2 years

We just started the #TRAIT2022 workshop at #CHI2022 ! Turns out our keynote speaker, John Lee @Jdlee888 wrote his keynote's abstract with with AI assistance (blue indicates contributions by a LLM)! And none us knew until he just told us now!! Amazing!!

0

1

19

Gagan Bansal

@bansalg_

7 months

Nice example of an online multi-agent playground built using #AutoGen !

0

4

17

Gagan Bansal

@bansalg_

1 year

I had fun visiting an interacting with colleagues at UCSB! Here are papers I discussed: 1. Modeling users: 2. Communicating Uncertainty: 3. Metrics: Data/code:

GitHub - microsoft/coderec_programming_states: Code and Data for: Reading Between the Lines:...

Code and Data for: Reading Between the Lines: Modeling User Behavior and Costs in AI-Assisted Programming - microsoft/coderec_programming_states

github.com

Alfonso Amayuelas

@AlfonAmayuelas

1 year

Las week, we had the @ucsbmmi Summit @ UCSB. It was great to listen to @bansalg_ from @MSFTResearch and understand how Copilot is changing the way we code. Users spend 50% of their time interacting with it and 20% verifying suggestions. Coders are 2x faster!

0

1

14

1

16

Gagan Bansal

@bansalg_

3 months

More full-time opportunities to work with our team!

Saleema Amershi

@SaleemaAmershi

3 months

📢📢More opportunities on our team at #MicrosoftResearch ! 📢📢 Now hiring for Senior and Principal level Researchers and Software Engineers. If you want to advance the Frontiers of #AI to empower people and AI agents to solve real-world problems, apply below! 👇 Please RT

1

26

78

1

2

16

Gagan Bansal

@bansalg_

2 years

There is still time to submit to the TRAIT workshop at #CHI2022 !!! Submissions are due on Feb 11th.

Gagan Bansal

@bansalg_

2 years

🚨Announcing #CHI2022 workshop on Trust and Reliance in AI-Human Teams (TRAIT) ft. keynote by @jdlee888 , panel w/ #HCI & #AI experts, group activities & more! w/ @alison_m_smith , @ZanaBucinca , @tongshuangwu , @d19fe8 , @JessicaHullman , @DrSimoneStumpf See:

3

30

123

0

3

16

Gagan Bansal

@bansalg_

7 months

If you are passionate about human-AI interaction and on the job market this year, we are hiring a full-time researcher🧑‍🦱+🤖 See the job post for details below.

Saleema Amershi

@SaleemaAmershi

7 months

📢📢We're hiring!📢📢 If you want to shape the future of AI and empower people and AI agents to collaboratively solve real-world problems, apply here: See also below for more exciting opportunities in #AI at #MSR with our partner teams. 👇👇

7

68

350

0

2

16

Gagan Bansal

@bansalg_

2 years

@JessicaHullman moderates a super interesting panel to evaluate our progress and envision the future of trust and reliance in human-AI Teams with leading AI and HCI researchers @mariadearteaga @SaleemaAmershi @kgajos @tmiller_unimelb @sigchi #TRAIT2022 #CHI2022

0

4

15

Gagan Bansal

@bansalg_

4 years

In fact, explanations increased the chance that users will accept its recommendation REGARDLESS of its correctness. Such systems seem deeply unsatisfying and fraught with ethical issues.(5/6)

1

0

13

Gagan Bansal

@bansalg_

4 months

New tool for benchmarking agents -- AutoGenBench! By @adamfourney , @qingyun_wu , @pyautogen

AutoGenBench -- A Tool for Measuring and Evaluating AutoGen Agents | AutoGen

AutoGenBench

microsoft.github.io

Besmira Nushi 💙💛

@besanushi

4 months

New blog by @adamfourney and @qingyun_wu on measurement tools for complex multi agent workflows in @pyautogen . AutoGenBench is a command line tool on pypi which handles downloading, configuring, running, and reporting supported benchmarks in AutoGen.➡️

1

6

28

0

12

Gagan Bansal

@bansalg_

2 months

If you work on human-AI interaction and agents, you might find the abstractions introduced in chapter 3 of the new #AutoGen tutorial practical and interesting 👇

AutoGen

@pyautogen

2 months

🚨We just released a new #AutoGen tutorial And with that getting started became even easier! First 5 chapter are already online to help you learn about - agents that can converse - termination - adding humans in-the-loop - code executors - multi-agent patterns Lets us know if

2

38

183

0

2

12

Gagan Bansal

@bansalg_

7 months

Checkout our new framework for building LLM agents! #AutoGen is already open source and growing very rapidly on Github. You can start using it today! More details soon… #LLMs #Microsoft #AI 👇👇👇

Chi Wang

@Chi_Wang_

7 months

Imagine if ✨multiple✨ ChatGPT agents could collaborate to solve complex tasks for you! 🧑‍🦰🤝🤖🤖🤖 📢 AutoGen: A new framework for building multi-agent LLM applications Repo: Stay tuned for a new AutoGen tech report on 10/5… #AutoGen #AI #LLMs #ML

7

45

163

0

2

11

Gagan Bansal

@bansalg_

3 years

Aptly put by @JessicaHullman -- "So the relationship more explanation = more [appropriate] trust should not be assumed when trust is mentioned as in the NIST report, just like it shouldn’t be assumed that more expression of uncertainty = more [appropriate] trust."

Jessica Hullman

@JessicaHullman

3 years

On NIST principles for explainable AI, and what's similar about these challenges and those in expressing uncertainty in model predictions. I see a lot of parallels despite the big difference in how much hype each gets

1

5

34

0

1

10

Gagan Bansal

@bansalg_

3 years

Thankful for colleagues at @MSFTResearch , #MicrosoftAether , and @uwcse for their commitment to develop reliable people-facing #AI systems! See announcement and link to a new open-source repository on backwards compatible #ML ⬇️

Eric Horvitz

@erichorvitz

3 years

Excited to share code for studying backward compatibility of #ML models—understanding changes in errors w/ model updates. Fun collab w/ @besanushi @bansalg_ @megha_byte @ecekamar @sytelus @dsweld @DeanCarignan & #MicrosoftAether eng team @MSFTResearch @NeurIPSConf #responsibleAI

0

1

14

1

2

10

Gagan Bansal

@bansalg_

2 years

More intense brainstorming sessions with larger groups at #TRAIT2022 #CHI2022

0

2

10

Gagan Bansal

@bansalg_

1 year

I remember finding @jennwvaughan 's advice really useful when attending conferences! Her point #8 "One new friend will often lead to many" is still my favorite!

Jenn Wortman Vaughan

@jennwvaughan

1 year

Attending #NeurIPS for the first time? It's been a while since I wrote this, but it's still relevant...

0

9

65

0

1

9

Gagan Bansal

@bansalg_

2 years

We've extended the submission deadline for #CHI2022 TRAIT workshop to Feb 24th!

Gagan Bansal

@bansalg_

2 years

There is still time to submit to the TRAIT workshop at #CHI2022 !!! Submissions are due on Feb 11th.

0

3

16

0

7

9

Gagan Bansal

@bansalg_

7 months

I am already mind blown by the reception of #AutoGen by OSS community! But I am also super excited about the numerous human-AI interaction questions that show up when users interact with and use multiple #LLM agents for their tasks...

AutoGen

@pyautogen

7 months

Join our rapidly growing community!🚀🚀🚀

0

1

15

0

9

Gagan Bansal

@bansalg_

3 years

Even expert users are susceptible to inappropriate reliance on AI advice 😮

Marzyeh

@MarzyehGhassemi

3 years

Are radiologists and IM/EM docs more susceptible to incorrect radiology advice when its "from an AI"? Our new paper "Do as AI Say" highlights the potential danger of human/AI advice anchoring. Blog post by author @harini824 !

6

53

228

0

8

Gagan Bansal

@bansalg_

6 months

More internships with our close collaborators:

Julia Kiseleva

@julia_kiseleva

6 months

🚀 Exciting summer internship opportunity for PhD students at @MSFTResearch ! Dive into innovative projects like #Orca and @pyautogen offering thrilling research challenges. Ready to be part of groundbreaking AI work? Apply here: #AutoGen #LLMs #AgentEval

3

21

101

2

1

8

Gagan Bansal

@bansalg_

3 years

This was one of the most well-organized workshop I've attended!

Upol Ehsan

@UpolEhsan

3 years

Many of you couldn't join us at the #HCXAI workshop at #CHI2021 . We received tons of requests to make the videos available online. We always want to broaden participation. This is for you. 🎁

4

17

70

1

0

8

Gagan Bansal

@bansalg_

3 years

Another great example of how AI systems are brittle and can fail in unexpected ways, and the need for human agency, control, and feedback in people-facing systems.

0

8

Gagan Bansal

@bansalg_

4 years

Finally caught up with exciting papers on explainable AI at #ICML2020 's Workshop on Human Interpretability #WHI2020 . Here are subset of many papers I liked and a TLDR: (1/4)

1

8

Gagan Bansal

@bansalg_

2 months

Highly recommend for any dev or researcher working on AI agents!

0

7

Gagan Bansal

@bansalg_

3 years

@peterbhase @__Owen___ Very nice resource! You may be interested in our studies from last summer that show that NONE of the prior works (except for one very recent study on open-domain QA) has observed "complementary" performance from explanations!

Gagan Bansal

@bansalg_

4 years

New work on explainable AI! w/ @tongshuangwu , J. Zhu, R. Fok, @besanushi , @ecekamar , M. Ribeiro, and @dsweld . When AIs advise people, does an explanation of its reasoning actually help the person? Does it let the human outperform the AI? Does it ...(1/6)

6

43

217

1

2

6

Gagan Bansal

@bansalg_

3 years

@tmiller_unimelb Here's a very recent counter-example/-domain though: For the task of open-domain QA explanations work and help end-users. Led by @AnaValeriaGlez @sriniiyer88 @robinomial @YasharMehdad See:

0

6

Gagan Bansal

@bansalg_

1 year

The workshop deadline is approaching soon— Feb 23rd!! #chi2023

Ken Holstein

@d19fe8

1 year

There's still time to submit to the CHI TRAIT workshop on Trust and Reliance in AI-Assisted Tasks! We welcome submissions from both researchers (Research Track) and practitioners (Industry Track). Submissions are due next Thursday, Feb 23 (AoE). CfP:

1

8

19

0

5

Gagan Bansal

@bansalg_

3 years

"Heatmaps often only highlight dog faces regardless of whether AI is correct or wrong."

Anh (Totti) Nguyen

@anh_ng8

3 years

In our study of fine-grained dog classification (🦮 / 🐩 / 🐕), human-AI teams where humans use heatmaps performed even worse than the AI alone. Heatmaps often only highlight dog faces regardless of whether AI is correct or wrong.

1

2

4

0

6

Gagan Bansal

@bansalg_

4 years

In search of complementary performance, we conducted new studies where human and AI performance was comparable. While we observed benefits from AI augmentation, they were NOT increased by showing state-of-art explanations. (4/6)

1

0

6

Gagan Bansal

@bansalg_

3 years

@annargrs @uwnlp @nlpnoah Is there a link to a recording of this talk? Thanks :-)

1

0

5

Gagan Bansal

@bansalg_

2 years

@roboticwrestler

0

5

Gagan Bansal

@bansalg_

2 months

Custom environments play a pivotal role in building useful agents! Checkout this new feature in #AutoGen !

AutoGen

@pyautogen

2 months

AutoGen's code execution capabilities have gotten an upgrade! You can use a Jupyter kernel to maintain a stateful session for code execution. 🤖🌎🔧 Learn more here:

1

7

36

0

5

Gagan Bansal

@bansalg_

4 years

...work better than simply displaying the AI’s confidence? In the quest for augmented intelligence, such questions deserve critical attention! (2/6)

1

0

5

Gagan Bansal

@bansalg_

4 years

While our novel Adaptive explanations showed promise, we must develop explanation algorithms and interfaces that lead to complementary performance, e.g., by enabling appropriate reliance, and providing significant value over simple baselines such as showing AI confidence. (6/6)

1

5

Gagan Bansal

@bansalg_

6 months

Massive update to AutoGen! It now supports gpt-4-vision! #autogen #gpt4V

Chi Wang

@Chi_Wang_

6 months

🚀 @pyautogen new release is here with gpt-4-vision-preview multimodal models support! 🛠️ Codebase updates for supporting openai-python v1. 📊 New unstructured data support in RAG & async features for get_human_input. 🔧 Fresh tools & improved docs for devs. #GPT4V #AI #AutoGen

0

16

51

0

5

Gagan Bansal

@bansalg_

3 years

Another instance of how imperfect AI in people-facing interfaces can cause real harm! And why human-centered AI and interfaces are necessary!

Maarten Sap (he/him)

@MaartenSap

3 years

This is why we need to keep scrutinizing the fairness of toxic or inappropriate content filters, and always let users circumvent the automatic systems.

0

3

16

0

5

Gagan Bansal

@bansalg_

3 years

@MaartenvSmeden @tongshuangwu @zacharylipton Sadly true. Especially #7 . And this is why user evaluations are crucial to get a realistic picture of current state of XAI. See:

Gagan Bansal

@bansalg_

4 years

New work on explainable AI! w/ @tongshuangwu , J. Zhu, R. Fok, @besanushi , @ecekamar , M. Ribeiro, and @dsweld . When AIs advise people, does an explanation of its reasoning actually help the person? Does it let the human outperform the AI? Does it ...(1/6)

6

43

217

0

5

Gagan Bansal

@bansalg_

3 months

@tongshuangwu @dsweld @cmuhcii I see @mynkgoel

1

0

5

Gagan Bansal

@bansalg_

1 year

Pretty cool feature for reading papers faster! CC: @EricTopol whose highlights on COVID related papers I find very useful.

Semantic Scholar

@SemanticScholar

1 year

🚨 New beta feature live! Do you skim through papers trying to get a glimpse in a minute? By turning on #Skimming in Semantic Reader, you can skim faster with automatically highlighted overlays of the key points. Now available for 9k papers on desktop!

0

13

28

1

4

Gagan Bansal

@bansalg_

2 years

...as we suggested in our call to arms paper And as we showed in our studies with open-domain QA, where explanations significantly improved appropriate reliance [2/2]

0

4

Gagan Bansal

@bansalg_

4 years

Prior work on XAI only considers the case when AI by itself was more accurate than the human and the human-AI team. Explanations raised team performance closer to AI but if accuracy were the sole objective, removing people would have performed even better in their settings! (3/6)

1

0

4

Gagan Bansal

@bansalg_

2 months

More opportunities to work with our lab!

Ahmed Awadallah

@AhmedHAwadallah

2 months

We are hiring senior and principal researchers and engineers to work on generative AI technologies including foundation models, small models and learning agent platforms. Applications at: and

2

21

72

0

1

4

Gagan Bansal

@bansalg_

7 months

Checkout this cool coverage of our work on #AutoGen

VentureBeat

@VentureBeat

8 months

Microsoft’s AutoGen framework allows multiple AI agents to talk to each other and complete your tasks

3

9

13

0

4

Gagan Bansal

@bansalg_

5 months

We’ll be again at the Microsoft booth tomorrow 2:30-3:00 pm CST!

Gagan Bansal

@bansalg_

5 months

Come learn more about agents and AutoGen at the Microsoft booth at #NeurIPS2023 ! @Chi_Wang_ , @qingyun_wu , and I will be there on Monday 12/11, between 9 am-noon CST and 3:30-4:00 pm CST. Location: Booth 1003 - Next to entrance Hall D cc: @pyautogen , @MSFTResearch

1

4

26

0

3

Gagan Bansal

@bansalg_

5 months

cc: @SaleemaAmershi @adamfourney @vykthur @ecekamar

0

4

Gagan Bansal

@bansalg_

4 years

We show that approaches maximizing AI accuracy (by using Log-loss) may lead to suboptimal team utility. Instead, we propose and optimize a new loss function based on the team's expected utility. (2/5)

1

0

2

Gagan Bansal

@bansalg_

4 years

cc: Joyce Zhou @cephcyn , my excellent co-author who is finally on Twitter. I misspelled their name in my original tweet--- so sorry, Joyce!

0

3

Gagan Bansal

@bansalg_

2 years

@jeffrey_heer @uwcse The course content looks amazing!

0

3

Gagan Bansal

@bansalg_

3 years

“5 to 20 fold increase in snares captured!” 😮

Toby Walsh

@TobyWalsh

3 years

Brilliant video asking you to rethink AI, featuring my amazing colleague @MilindTambe_AI and some endangered elephants @thebrilliantHQ via @YouTube

2

9

41

0

3

Gagan Bansal

@bansalg_

6 months

Make multiple GPTs collaborate with AutoGen!!

AutoGen

@pyautogen

6 months

We just released a new version of #AutoGen and added compatibility with #OpenAI Assistants! This means you can now make multiple GPTs collaborate to solve complex tasks 🤖🤖🤖 Checkout our new blog post for details:

6

26

116

0

3

Gagan Bansal

@bansalg_

10 months

@pranavrajpurkar Pranav, you may be interested in our 2021 CHI paper where related concepts of complementary performance and appropriate reliance: @tongshuangwu @dsweld

Does the Whole Exceed its Parts? The Effect of AI Explanations on...

Many researchers motivate explainable AI with studies showing that human-AI team performance on decision-making tasks improves when the AI explains its recommendations. However, prior studies...

arxiv.org

0

3

Gagan Bansal

@bansalg_

3 years

@leavittron @arimorcos @MLRetrospective Especially agree with "under-utilization of user research for human verification." Though I'd say the focus on human subject studies is increasingly rapidly! You may find our recent work relevant :)

Gagan Bansal

@bansalg_

4 years

New work on explainable AI! w/ @tongshuangwu , J. Zhu, R. Fok, @besanushi , @ecekamar , M. Ribeiro, and @dsweld . When AIs advise people, does an explanation of its reasoning actually help the person? Does it let the human outperform the AI? Does it ...(1/6)

6

43

217

1

0

3

Gagan Bansal

@bansalg_

2 years

What @tmiller_unimelb said! Nice to see more and more researchers and domains ask one of the most important questions in the context of #XAI and #AI in general.

Tim Miller

@tmiller_uq

2 years

If I could like this more than once, I would.

0

1

9

0

3

Gagan Bansal

@bansalg_

3 months

Super cool article!

Eddie Aftandilian

@eaftandilian

3 months

Check out our CACM article on measuring the impact of GitHub Copilot on developer productivity! There’s a brief video here as well:

1

7

23

0

3

Gagan Bansal

@bansalg_

2 years

To learn more about our team, visit

HAX Team - Microsoft Research

Creating the next generation of AI-based systems and experiences The Human-AI eXperiences (HAX) team is part of Microsoft Research AI Frontiers. We bring together leaders in human-computer interact...

www.microsoft.com

0

1

Gagan Bansal

@bansalg_

6 months

🙌

Chi Wang

@Chi_Wang_

6 months

#autogen is mentioned by Satya around 14:00 @pyautogen #AI

3

6

34

0

3

Gagan Bansal

@bansalg_

4 years

@imjuhokim @minsuk_chang @aliceoh @roboticwrestler @nekozzang Congrats Dr. @minsuk_chang !

0

2

Gagan Bansal

@bansalg_

1 year

@kous2v So sorry Koustuv, I can’t even imagine how frustrating and disruptive this must be :(

1

0

2

Gagan Bansal

@bansalg_

4 years

@laura_rieger_de @tongshuangwu @besanushi @ecekamar @dsweld Thank you! We tested w/ non-experts (MTurk), but even w/ experts, deployers should test & ensure explanations don't exacerbate inappropriate reliance or conf. bias. @ihsgnef 's work shows instances where experts are more immune to bad system suggestions:

0

2

Gagan Bansal

@bansalg_

7 months

The new CMD + I feature in @github #CoPilot is pretty darn neat! 🤌🤌

0

2

Gagan Bansal

@bansalg_

3 years

@AndrewLBeam @MarzyehGhassemi @DrLukeOR Especially agree with "We should advocate for thorough ..validation of these systems..., showing that patient and health-care outcomes are improved" Precisely why we argued for carefully measuring effect of explanations on human-AI team performance:

Gagan Bansal

@bansalg_

4 years

New work on explainable AI! w/ @tongshuangwu , J. Zhu, R. Fok, @besanushi , @ecekamar , M. Ribeiro, and @dsweld . When AIs advise people, does an explanation of its reasoning actually help the person? Does it let the human outperform the AI? Does it ...(1/6)

6

43

217

2

0

2

Gagan Bansal

@bansalg_

3 years

"Anthropic said its work would be focused on 'large-scale AI models', including making the systems more easy to interpret and 'building ways to more tightly integrate human feedback into the development and deployment of these systems'."

Subbarao Kambhampati (కంభంపాటి సుబ్బారావు)

@rao2z

3 years

The cup of open #AI runneth over. There is the used to be open AI, there is the wannabe open AI, and now apparently there is the really like-fer-sure-this-time wannabe real open AI..

1

3

4

0

2

Gagan Bansal

@bansalg_

2 months

Among many inspiring feedback we received from this expert, who Ive admired for more than a decade, it was so fascinating to hear how much @pyautogen has spurred bottom-up creativity and advanced our worlds understanding of AI agents! cc: @ekzhu @jack_gerrits

Chi Wang

@Chi_Wang_

2 months

Had a conversation with an iconic leader + my mentor^2 and was told that he was a fan of #AutoGen ! That made my day❤️‍🔥 Super inspired by an insight to the uniqueness of #AutoGen 🦄

2

1

21

0

2

Gagan Bansal

@bansalg_

4 months

@mishumausam @dsweld @etzioni @pmddomingos @fox_dieter17849 @rao2z Congratulations!

0

Gagan Bansal

@bansalg_

2 years

@AlexKale17 Hopefully there will be another one 😉

0

2

Gagan Bansal

@bansalg_

2 years

Pretty cool and impactful result!

Stefan

@stefanjwojcik

2 years

The Birdwatch algorithm surfaces notes to potentially misleading Tweets. Using survey data, we find notes selected by the algorithm reduce the likelihood of agreeing with the substance of a potentially misleading Tweet by about 26%.

1

0

4

0

2

Gagan Bansal

@bansalg_

4 years

Team-loss accounts for people adjusting their trust in AI based on the stakes and the cost of human effort. Positive effects can be observed in both synthetic and real datasets and the shift in behavior reflects the encoded human-centered properties. (3/5)

1

0

2

Gagan Bansal

@bansalg_

2 years

@HsseinMzannar @adamfourney It was so great to have you and your expertise in our collaboration, Hussein!

0

2

Gagan Bansal

@bansalg_

2 years

cc @alison_m_smith @ZanaBucinca @tongshuangwu @d19fe8 @JessicaHullman @DrSimoneStumpf

1

0

2

Gagan Bansal

@bansalg_

10 months

@besanushi @tdietterich @GaryMarcus @xiao_ted @GoogleDeepMind Backwards compatibility for foundational models would be so helpful! But I think end applications can also build in some of this robustness!

1

0

2

Gagan Bansal

@bansalg_

6 months

More full time positions with Frontiers:

Saleema Amershi

@SaleemaAmershi

7 months

📢📢We're hiring!📢📢 If you want to shape the future of AI and empower people and AI agents to collaboratively solve real-world problems, apply here: See also below for more exciting opportunities in #AI at #MSR with our partner teams. 👇👇

7

68

350

0

2

Gagan Bansal

@bansalg_

6 months

Haha clever and meta! It's built with AutoGen to help build with AutoGen using discussions from people building with AutoGen! 🤓 🤪

AW

@TrainedOnTest

6 months

Can't find the answers in the docs? no problem 🤖 I built a multi-agent LLM application to query the collective developer knowledge of the @pyautogen discord server message history. @chainlit_io @OpenAI @trychroma

2

5

17

0

1

2

Gagan Bansal

@bansalg_

6 months

Checkout new blog post on AutoGen and Semantic Kernel by @johnmaeda and Devis Lucato!

Alex Chao

@alexchaomander

6 months

🤝 AutoGen + Semantic Kernel! Devis Lucato shows off how AutoGen can be the basis of a new planner that you can use in your Semantic Kernel applications, unlocking a whole class of interesting scenarios because of conversational multi-agents!

3

15

50

0

2