Vaidehi Patil
@vaidehi_patil_
PhD student @unccs @uncnlp, advised by @mohitban47 | Undergrad @IITBombay | Intern @GoogleDeepMind; Prev. Intern @Apple @AmazonScience @AdobeResearch
🚨 Introducing our @TmlrOrg paper “Unlearning Sensitive Information in Multimodal LLMs: Benchmark and Attack-Defense Evaluation” W:nt UnLOK-VQA, a benchmark to evaluate unlearning in vision-and-language models—where both images and text may encode sensitive or private…
Life update: I’m excited to share that I’ll be starting as faculty at the Max Planck Institute for Software Systems(@mpi_sws_) this Fall!🎉 I’ll be recruiting PhD students in the upcoming cycle, as well as research interns throughout the year: lasharavichander.github.io/contact.html
🥳 Gap year update: I'll be joining @allen_ai/@UW for 1 year (Sep2025-Jul2026 -> @JHUCompSci) & looking forward to working with amazing folks there, incl. @RanjayKrishna, @HannaHajishirzi, Ali Farhadi. 🚨 I’ll also be recruiting PhD students for my group at @JHUCompSci for Fall…
Sharing some personal updates 🥳: - I've completed my PhD at @unccs! 🎓 - Starting Fall 2026, I'll be joining the Computer Science dept. at Johns Hopkins University (@JHUCompSci) as an Assistant Professor 💙 - Currently exploring options + finalizing the plan for my gap year (Aug…
🎉 Our paper, GenerationPrograms, which proposes a modular framework for attributable text generation, has been accepted to @COLM_conf! GenerationPrograms produces a program that executes to text, providing an auditable trace of how the text was generated and major gains on…
Excited to share GenerationPrograms! 🚀 How do we get LLMs to cite their sources? GenerationPrograms is attributable by design, producing a program that executes text w/ a trace of how the text was generated! Gains of up to +39 Attribution F1 and eliminates uncited sentences,…
At ICML? Interested in how we can do more with interpretability to have practical impact on the rest of AI? Come to our workshop this Saturday!
Hope everyone’s getting the most out of #icml25. We’re excited and ready for the Actionable Interpretability (@ActInterp) workshop this Saturday! Check out the schedule and join us to discuss how we can move interpretability toward more practical impact.
I’ll be at #ICML2025 this week to present ScPO: 📌 Wednesday, July 16th, 11:00 AM-1:30 PM 📍East Exhibition Hall A-B, E-2404 Stop by or reach out to chat about improving reasoning in LLMs, self-training, or just tips about being on the job market next cycle! 😃
🚨 Self-Consistency Preference Optimization (ScPO)🚨 - New self-training method without human labels - learn to make the model more consistent! - Works well for reasoning tasks where RMs fail to evaluate correctness. - Close to performance of supervised methods *without* labels,…
If you're attending @icmlconf, join us for the exciting Workshop on Machine Unlearning for Generative AI (MUGen)!
🚨Exciting @icmlconf workshop alert 🚨 We’re thrilled to announce the #ICML2025 Workshop on Machine Unlearning for Generative AI (MUGen)! ⚡Join us in Vancouver this July to dive into cutting-edge research on unlearning in generative AI—featuring an incredible lineup of…
AI text-to-speech programs could “unlearn” how to imitate certain people trib.al/Dr8TEmb
Overdue job update -- I am now: - A Visiting Scientist at @schmidtsciences, supporting AI safety and interpretability - A Visiting Researcher at the Stanford NLP Group, working with @ChrisGPotts I am so grateful I get to keep working in this fascinating and essential area, and…
🚨 Excited to attend #ICML2025 and share our latest work (@OptML_MSU) on LLM unlearning -- think of it as AI surgery: removing harmful knowledge while preserving general utility. Catch us at: 🔹 [Paper 1] Tues, July 15 @ 4:30pm PT | E-1108 📄 Invariance Makes LLM Unlearning…
🥳 Excited to share our work -- Retrieval-Augmented Generation with Conflicting Evidence -- on addressing conflict in RAG due to ambiguity, misinformation, and noisy/irrelevant evidence has been accepted to @COLM_conf #COLM2025! Our new benchmark RAMDocs proves challenging for…
🚨Real-world retrieval is messy: queries can be ambiguous, or documents may conflict/have incorrect/irrelevant info. How can we jointly address all these problems? We introduce: ➡️ RAMDocs, a challenging dataset with ambiguity, misinformation, and noise. ➡️ MADAM-RAG, a…
I've officially joined Meta Superintelligence Labs (MSL) org in the Bay Area. I'll be working on critical aspects of pre-training, synthetic data and RL for the next generation of models. Humbled and eager to contribute to the quest for superintelligence. @AIatMeta
🎉 Excited to share that TaCQ (Task-Circuit Quantization), our work on knowledge-informed mixed-precision quantization, has been accepted to #COLM2025 @COLM_conf! Happy to see that TaCQ was recognized with high scores and a nice shoutout from the AC – big thanks to @EliasEskin…
🚨Announcing TaCQ 🚨 a new mixed-precision quantization method that identifies critical weights to preserve. We integrate key ideas from circuit discovery, model editing, and input attribution to improve low-bit quant., w/ 96% 16-bit acc. at 3.1 avg bits (~6x compression)…
🥳Our work UTGen & UTDebug on teaching LLMs to generate effective unit tests & improve code debugging/generation has been accepted to @COLM_conf #COLM2025! Stay tuned for more exciting results -- e.g., using 32B-scale UTGen models to improve debugging with frontier models like…
🚨 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 improve LLM-based code debugging by addressing 3 key…
🥳 Excited to share that I’ll be joining the CS Department at UNC-Chapel Hill (@unccs @unc_ai_group) as an Assistant Professor starting Fall 2026! Before that, I’ll be working at Ai2 Prior (@allen_ai @Ai2Prior) and UW (@uwcse) on multimodal understanding and generation.
“How will my model behave if I change the training data?” Recent(-ish) work w/ @logan_engstrom: we nearly *perfectly* predict ML model behavior as a function of training data, saturating benchmarks for this problem (called “data attribution”).
🎉Excited to announce VEEGIE has been accepted to #ICCV2025 ! VEGGIE is a unified MLLM + Diffusion framework for instructional video editing. It presents a systematic approach spanning data, model, benchmark, and evaluation design, and shows strong multi-skill editing +…
🚨 Excited to announce MF2, a new+challenging long-video understanding dataset! MF2 covers open-license movies and focuses on key events/arcs/causal chains in the film. While people can answer MF2 questions easily, even the strongest models like Gemini 2.5 pro struggle with it!…
🚨Meet MF²: Movie Facts & Fibs: a new benchmark for long-movie understanding! 🤔Do you think your model understands movies? Unlike existing benchmarks, MF² targets memorable events, emotional arcs 💔, and causal chains 🔗 — things humans recall easily, but even top models like…
🚨 Announcing CINGS, a new method for improving grounding in LLMs and VLMs! CINGS works at the instruction-tuning stage, teaching models to incorporate contextual info instead of over-relying on parametric knowledge. Gains in both text and multimodal settings, and nice…
🚨 Want models to better utilize and ground on the provided knowledge? We introduce Context-INformed Grounding Supervision (CINGS)! Training LLM with CINGS significantly boosts grounding abilities in both text and vision-language models compared to standard instruction tuning.