Lightnews — Scholar-powered news

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · May 20

🔥 Huge CONGRATS to Jaemin + @jhucompsci.bsky.social! 🎉

Very proud of his journey as an amazing researcher (covering groundbreaking, foundational research on important aspects of multimodality+other areas) & as an awesome, selfless mentor/teamplayer 💙
-- Apply to his group & grab him for gap year!

Jaemin Cho @jmincho.bsky.social · May 20

Some personal updates:
- I've completed my PhD at @unccs.bsky.social! 🎓
- Starting Fall 2026, I'll be joining the CS dept. at Johns Hopkins University @jhucompsci.bsky.social as an Assistant Professor 💙
- Currently exploring options for my gap year (Aug 2025 - Jul 2026), so feel free to reach out! 🔎

1 1 6

Reposted by Zaid Khan

Jaemin Cho @jmincho.bsky.social · May 20

Some personal updates:
- I've completed my PhD at @unccs.bsky.social! 🎓
- Starting Fall 2026, I'll be joining the CS dept. at Johns Hopkins University @jhucompsci.bsky.social as an Assistant Professor 💙
- Currently exploring options for my gap year (Aug 2025 - Jul 2026), so feel free to reach out! 🔎

3 5 26

Reposted by Zaid Khan

Vaidehi Patil @vaidehipatil.bsky.social · May 7

🚨 Introducing our @tmlrorg.bsky.social paper “Unlearning Sensitive Information in Multimodal LLMs: Benchmark and Attack-Defense Evaluation”
We present UnLOK-VQA, a benchmark to evaluate unlearning in vision-and-language models, where both images and text may encode sensitive or private information.

1 8 10

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · May 5

🔥 BIG CONGRATS to Elias (and UT Austin)! Really proud of you -- it has been a complete pleasure to work with Elias and see him grow into a strong PI on *all* axes 🤗

Make sure to apply for your PhD with him -- he is an amazing advisor and person! 💙

Elias Stengel-Eskin @esteng.bsky.social · May 5

Extremely excited to announce that I will be joining
@utaustin.bsky.social Computer Science in August 2025 as an Assistant Professor! 🎉

1 4 11

Reposted by Zaid Khan

Elias Stengel-Eskin @esteng.bsky.social · May 5

Extremely excited to announce that I will be joining
@utaustin.bsky.social Computer Science in August 2025 as an Assistant Professor! 🎉

5 9 43

Reposted by Zaid Khan

Elias Stengel-Eskin @esteng.bsky.social · Apr 29

✈️ Heading to #NAACL2025 to present 3 main conf. papers, covering training LLMs to balance accepting and rejecting persuasion, multi-agent refinement for more faithful generation, and adaptively addressing varying knowledge conflict.

Reach out if you want to chat!

1 5 15

Reposted by Zaid Khan

Elias Stengel-Eskin @esteng.bsky.social · Apr 24

Check out 🚨CAPTURe🚨 -- a new benchmark testing spatial reasoning by making VLMs count objects under occlusion.

SOTA VLMs (GPT-4o, Qwen2-VL, Intern-VL2) have high error rates on CAPTURe (but humans have low error ✅) and models struggle to reason about occluded objects.

arxiv.org/abs/2504.15485

🧵👇

1 4 5

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · Apr 21

In Singapore for #ICLR2025 this week to present papers + keynotes 👇, and looking forward to seeing everyone -- happy to chat about research, or faculty+postdoc+phd positions, or simply hanging out (feel free to ping)! 🙂

Also meet our awesome students/postdocs/collaborators presenting their work.

1 4 19

Reposted by Zaid Khan

Archiki Prasad @archiki.bsky.social · Apr 18

🚨Real-world retrieval is messy: queries are ambiguous or docs conflict & have incorrect/irrelevant info. How can we jointly address these problems?

➡️RAMDocs: challenging dataset w/ ambiguity, misinformation & noise
➡️MADAM-RAG: multi-agent framework, debates & aggregates evidence across sources

🧵⬇️

3 7 14

Zaid Khan @codezakh.bsky.social · Apr 15

It was a fun collaboration with @esteng.bsky.social @archiki.bsky.social @jmincho.bsky.social @mohitbansal.bsky.social! 🥳

Paper: arxiv.org/abs/2504.09763
Project Page: zaidkhan.me/EFAGen
Datasets + Models: huggingface.co/collections/...
HF Paper: huggingface.co/papers/2504....

Executable Functional Abstractions: Inferring Generative Programs for Advanced Math Problems

Scientists often infer abstract procedures from specific instances of problems and use the abstractions to generate new, related instances. For example, programs encoding the formal rules and properti...

arxiv.org

2

Zaid Khan @codezakh.bsky.social · Apr 15

EFAs can be used for adversarial search to find harder problem variants. This has some interesting potential uses, such as finding fresh problems for online RL or identifying gaps / inconsistencies in a model’s reasoning ability. We can find variants of even Level 1 problems (GPT-4o) solves wrong.

1

Zaid Khan @codezakh.bsky.social · Apr 15

EFAGen can infer EFAs for diverse sources of math data.

We demonstrate this by inferring EFAs on the NuminaMath dataset, which includes problems ranging from grade school to olympiad level problems. EFAGen can successfully infer EFAs for all math sources in NuminaMath, even olympiad-level problems.

1

Zaid Khan @codezakh.bsky.social · Apr 15

EFAs are effective at augmenting training data.

Getting high-quality math data is expensive. EFAGen offers a way to improve upon existing math training data by generating problem variants through EFAs. EFA-based augmentation leads to consistent improvements across all evaluation metrics.

1

Zaid Khan @codezakh.bsky.social · Apr 15

LMs can self-improve at inferring EFAs with execution feedback!

We self-train Llama-3.1-8B-Instruct with rejection finetuning using our derived unit tests as a verifiable reward signal and see substantial improvements in the model’s ability to infer EFAs, especially on harder problems.

1

Zaid Khan @codezakh.bsky.social · Apr 15

Key Insight💡: We formalize properties any valid EFA must possess as unit tests and treat EFA inference as a program synthesis task that we can apply test-time search to.

1

Zaid Khan @codezakh.bsky.social · Apr 15

➡️ EFAGen can generate data to augment static math datasets
➡️ EFAGen can infer EFAs for diverse + difficult math problems
➡️ Use EFAs to find + generate harder variants of existing math problems
➡️ LLMs can self-improve at writing EFAs

1

Zaid Khan @codezakh.bsky.social · Apr 15

What if we could transform advanced math problems into abstract programs that can generate endless, verifiable problem variants?

Presenting EFAGen, which automatically transforms static advanced math problems into their corresponding executable functional abstractions (EFAs).
🧵👇

1 5 15

Reposted by Zaid Khan

Archiki Prasad @archiki.bsky.social · Mar 27

🥳🥳 Honored and grateful to be awarded the 2025 Apple Scholars in AI/ML PhD Fellowship! ✨

Huge shoutout to my advisor @mohitbansal.bsky.social, & many thanks to my lab mates @unccs.bsky.social , past collaborators + internship advisors for their support ☺️🙏

machinelearning.apple.com/updates/appl...

1 3 14

Reposted by Zaid Khan

Vaidehi Patil @vaidehipatil.bsky.social · Feb 25

🚨 Introducing UPCORE, to balance deleting info from LLMs with keeping their other capabilities intact.

UPCORE selects a coreset of forget data, leading to a better trade-off across 2 datasets and 3 unlearning methods.

🧵👇

2 5 12

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · Feb 5

🚨 Check out "UTGen & UTDebug" for learning to automatically generate unit tests (i.e., discovering inputs which break your code) and then applying them to debug code with LLMs, with strong gains (>12% pass@1) across multiple models/datasets! (see details in 🧵👇)

1/4

Archiki Prasad @archiki.bsky.social · Feb 4

🚨 Excited to share: "Learning to Generate Unit Tests for Automated Debugging" 🚨
which introduces ✨UTGen and UTDebug✨ for teaching LLMs to generate unit tests (UTs) and debugging code from generated tests.

UTGen+UTDebug yields large gains in debugging (+12% pass@1) & addresses 3 key questions:

🧵👇

1 4 7

Reposted by Zaid Khan

Elias Stengel-Eskin @esteng.bsky.social · Feb 4

🚨 Excited to announce UTGen and UTDebug, where we first learn to generate unit tests and then apply them to debugging generated code with LLMs, with strong gains (+12% pass@1) on LLM-based debugging across multiple models/datasets via inf.-time scaling and cross-validation+backtracking!

🧵👇

Archiki Prasad @archiki.bsky.social · Feb 4

🚨 Excited to share: "Learning to Generate Unit Tests for Automated Debugging" 🚨
which introduces ✨UTGen and UTDebug✨ for teaching LLMs to generate unit tests (UTs) and debugging code from generated tests.

UTGen+UTDebug yields large gains in debugging (+12% pass@1) & addresses 3 key questions:

🧵👇

5 8

Reposted by Zaid Khan

Archiki Prasad @archiki.bsky.social · Feb 4

🚨 Excited to share: "Learning to Generate Unit Tests for Automated Debugging" 🚨
which introduces ✨UTGen and UTDebug✨ for teaching LLMs to generate unit tests (UTs) and debugging code from generated tests.

UTGen+UTDebug yields large gains in debugging (+12% pass@1) & addresses 3 key questions:

🧵👇

1 7 18

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · Jan 27

-- positional bias of faithfulness for long-form summarization
-- improving generation faithfulness via multi-agent collaboration

(PS. Also a big thanks to ACs+reviewers for their effort!)

1 2

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · Jan 27

-- safe T2I/T2V gener
-- generative infinite games
-- procedural+predictive video repres learning
-- bootstrapping VLN via self-refining data flywheel
-- automated preference data synthesis
-- diagnosing cultural bias of VLMs
-- adaptive decoding to balance contextual+parametric knowl conflicts
🧵

1 1 3

Reposted by Zaid Khan

Mohit Bansal @mohitbansal.bsky.social · Jan 27

-- adapting diverse ctrls to any diffusion model
-- balancing fast+slow sys-1.x planning
-- balancing agents' persuasion resistance+acceptance
-- multimodal compositional+modular video reasoning
-- reverse thinking for stronger LLM reasoning
-- lifelong multimodal instruc tuning via dyn data selec
🧵

1 1 1