Recipient of an ICML 2025 Outstanding Paper Award, CollabLLM improves how LLMs collaborate with users, including knowing when to ask questions and how to adapt tone and communication style to different situations. This approach helps move AI toward more user-centric and

🏆Thrilled that #CollabLLM won the #ICML2025 Outstanding Paper Award!. We propose a new approach to optimize human-AI collaboration, which is critical for agents. Congratulations to my fantastic co-authors; great job @ShirleyYXWu and Michel Galley driving the project!👏

📝 CollabLLM: From Passive Responders to Active Collaborators. Most LLMs provide helpful responses turn-by-turn but lack strategic foresight. This paper addresses a key limitation of existing LLMs: their inability to optimize for long-term collaboration. 1/5

Today's #PaperOfTheDay: ICML'25 Outstanding Paper #CollabLLM by Stanford × Microsoft.Stops guessing, starts GUIDING! 🔥.→ Trains models with Multiturn Reward (MR).→ Asks "Did this move conversation closer to the goal?".→ Simulates future turns ✅ Measures real impact.#bohrium

A new paper just dropped:.COLLABLLM: From Passive Responders to Active Collaborators

@icmlconf [4/9] How does CollabLLM work?.🤖<->👥 Model and User simulator create future conversations. 📊 Compute Multiturn-aware Rewards for the future conversations (e.g., task, efficiency, engagement) . 🔄Model learns to maximize long-term impact.→ Forward-looking LLM that

sn-news: #ai #ml #llm #hmi CollabLLM: Teaching LLMs to collaborate with users

CollabLLM teaches LLMs to collaborate with users

CollabLLM: Teaching LLMs to collaborate with users #machinelearning #ai.

CollabLLM: Teaching LLMs to collaborate with users

Gave a keynote at #TheWebConf2025 in Sydney! 🌏. Talked about advancing LLMs from retrieval to reasoning with:. 🧠 STaRK: semi-structured QA benchmark.🔧 AvaTaR: tool-using LLM agents.🤝 CollabLLM: models that collaborate.📊 POPPER: auto hypothesis testing.Slides:

🚨 BREAKING: Microsoft unveils CollabLLM, a game-changer in AI-human synergy! This new tech empowers LLMs to seamlessly partner with users, enhancing productivity and creativity. 🤝 Dive into the future of collaboration!. #AI #TechInnovation.

CollabLLM teaches LLMs to collaborate with users

CollabLLM: Teaching LLMs to collaborate with users - Microsoft Research #LLMs.

🤯 LLMs trained to *ask* questions? CollabLLM simulates convos & rewards responses that improve the *whole* dialogue, not just the next turn. User study saw higher quality docs & f.

Mind-blown by CollabLLM! 🚀 By training LLMs to ask clarifying questions & adapt to context, researchers achieved higher-quality outputs & faster task completion times. Imagine ap.

CollabLLM: Teaching LLMs to collaborate with users

@icmlconf [7/9] Ever chat with LLMs and get more annoying over time? 😤. We tracked user ratings every 3 turns,.📉 Base model: User ratings decline (users get frustrated) .📈 CollabLLM: User ratings maintain/increase!. CollabLLM gets better at collaborating the longer you chat ✨

CollabLLM: Teaching LLMs to collaborate with users.

CollabLLM: Teaching LLMs to collaborate with users - Microsoft Research