🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Excited to share our #HRI2025 paper “Social Robots as Social Proxies for Fostering Connection and Empathy Towards Humanity” 🧵(1/6) 📚Preprint: arxiv.org/abs/2502.00221

❓Can LLM agents generate personalized persuasive language than human experts while staying truthful? 🏡 We conducted an experiment with human home buyers and the answer is YES! Learning from Zillow real estate listings, our AI Realtor wins over human experts (Elo 1315 v 947)

New work from Akhila, my first PhD intern with Violet Peng!!

Presenting this work on Thursday at #HRI2025! Hope to see you there ☺️🇦🇺

🚨New Breakthrough in Tip-of-the-Tongue (TOT) Retrieval Research! We address data limitations and offer a fresh evaluation method for the TOT complex queries. Curious how TREC TOT track test queries are created? Check out this thread🧵 and our paper📄: arxiv.org/abs/2502.17776

Reward models for LMs are meant to align outputs with human preferences—but do they accidentally encode dialect biases? 🤔 Excited to share our paper on biases against African American Language in reward models, accepted to #NAACL2025 Findings! 🎉 arxiv.org/abs/2502.12858 (1/10)

"petition": Could we make the pre-recorded videos for underline optional (instead of required), for all *ACL conferences?

How to construct repo-level coding environments in a scalable way? Checkout RepoST: an automated framework to construct repo-level environments using Sandbox Testing (repost-code-gen.github.io) Models trained with RepoST data can generalize well to other datasets (e.g., RepoEval)

Thank you for inviting me! It was fun discussing what we’d need to setup better culturally contextual safety guardrails ~

These days RAG systems have gotten popular for boosting LLMs—but they're brittle💔. Minor shifts in phrasing (✍️ style, politeness, typos) can wreck the pipeline. Even advanced components don’t fix the issue. Check out this extensive eval by Neel Bhandari and Tianyu (Tiya) Cao!

Real user queries often look different from the clean, concise ones in academic benchmarks - ambiguity, full of typos, and much less readable. We show that even strong RAG systems quickly break under these conditions. Awesome project led by Neel Bhandari and Tianyu (Tiya) Cao!!

Check out PolyGuard 🤛 Our state-of-the-art safety moderation tool—now supporting 17 languages! Open source and built to make online spaces safer for everyone 🤩

Excited to share PolyGuard 🛡️, our new state-of-the-art multilingual safety detector. PolyGuard supports 17 languages and outperforms all open-source and commercial moderation tools!

When you interact with ChatGPT, have you wondered if they would ever "lie" to you? We found that in scenarios where truthfulness conflicts with achieving goals, LLMs often choose deception. Our new #NAACL2025 paper, "AI-LIEDAR ," reveals all models tested were truthful less than

📢 The SoLaR workshop will be collocated with COLM! Conference on Language Modeling SoLaR is a collaborative forum for researchers working on responsible development, deployment and use of language models. We welcome both technical and sociotechnical submissions, deadline July 5th!

🌏How culturally safe are large vision-language models? 👉LVLMs often miss the mark. We introduce CROSS, a benchmark of 1,284 image-query pairs across 16 countries & 14 languages, revealing how LVLMs violate cultural norms in context. ⚖️ Evaluation via CROSS-EVAL 🧨 Safety

I'm a featured interview in our latest behind-the-scenes release! We break down the ML and perception that drives the whole-body manipulation behaviors from last year. It starts with a neat demo of Atlas's range-of-motion and our vision foundation models. youtu.be/oe1dke3Cf7I?si…

✨ Unlock the power of synthetic data! Explore "Synthetic Data in the Era of LLMs" at #ACL2025NLP. This tutorial will build a shared understanding of recent progress, major methods, applications, and open problems in synthetic data generation for NLP. 2025.aclweb.org/program/tutori…

Yong Zheng-Xin (Yong) Cohere Labs Super interesting work! We address these gaps by releasing multilingual safety models (PolyGuard) along with an evaluation benchmark (PolyGuardPrompts) and large-scale training dataset (PolyGuardMix): x.com/kpriyanshu256/…