The NLP group at the University of Washington.
Seattle, WA
Joined September 2015
- Tweets1.3K
- Following187
- Followers15.1K
- Likes1.1K
UW NLP retweeted
Human annotations are getting more important as models scale, but they're expensive and still lack fine-grained signals important for learning‼️
We introduce ⭐️IsoVRBench⭐️, video pairs generated by targeted interventions without human labels, for reward model eval and training
UW NLP retweeted
Do coding agents need to finish a rollout and train on the complete trajectory? Not really.
We propose Contextual Early Reward (CER): predict the final reward from a partial trajectory, and use it for both test-time scaling and RL.
📄 arxiv.org/abs/2609.31995🧵👇
UW NLP retweeted
Evaluating self-improving agents means measuring not only what they can do, but how efficiently they get better through experience.
Huge thanks to my amazing collaborators and advisors: @anton_bakhtin @RulinShao @syhw @uralik1 @rob_fergus @prfsanjeevarora @KurtKeutzer @jaseweston @1_Anuj_M_1 @anirudhg9119, across @AIatMeta, @berkeley_ai, @uwnlp, and @Princeton.
Paper: arxiv.org/abs/2610.08902
Blog: harmandotpy.github.io/agent-…
Check out this fun project by @StellaLisy on learning to explore in image generation post-training. We enjoyed annotating the monkey penguins🐒🐧!
UW NLP retweeted
We introduce ExploreNet🔍 for learnable exploration in diffusion RL.
ExploreNet learns an adaptive exploration distribution conditioned on the current state, rewarded by the diversity of the rollouts.
➡️ Faster, more targeted learning in diffusion GRPO.
KV Cache cry no more.
Blog post by @RulinShao and @oscar_yinn
We recently released Context Language Models (CLMs), and got many questions on KV Cache.
Our latest blog explains
1). how we measure cache-ware FLOPs
2). how to efficiently serve CLM
wai-org.com/blog/clm/
UW NLP retweeted
Wrote a little blog with @RulinShao to visualize and explain some neat ways to reduce FLOPs usage for CLMs, check it out!!
We recently released Context Language Models (CLMs), and got many questions on KV Cache.
Our latest blog explains
1). how we measure cache-ware FLOPs
2). how to efficiently serve CLM
wai-org.com/blog/clm/
UW NLP retweeted
AI systems should be designed for AI. If prefix caching gets in the way of free context access, we can fix the caching.
This is the idea behind our Suffix Cache Reuse (SCR) for CLR serving. Check out our blogpost for details (fun visuals included)! wai-org.com/blog/clm/
Maybe the next step is to give CLMs direct agency over the cache itself. I’m excited to see what more creative and adaptive cache-aware context management could look like.
It's also a good time to follow our account @waiorg , a student org we created to share our work on agents. Currently covering: TMax, CLM.
UW NLP retweeted
‼️The Bitter Lesson for context management: Giving LMs unrestricted control over their context beats human-designed SOTA!
Introducing 🩵Context Language Models (CLMs)🩵
- Natively manage their own context
- Treat context as a file
- Learn policies in CLM weights, no harness
UW NLP retweeted
Image generation models CANNOT follow instructions (yet)🚨
We introduce Visual Verifiable Rewards (VVR) to solve it✅
VVRBench asks the model to "generate three yellow circles"🟡
Surprisingly, SOTA models, including GPT-Image-2.5-Sunburst, fail on complex VVR tasks.
RL with VVR (RLVVR), and mixing VVR intro standard image post-training rewards, improves instruction following in natural prompts.
UW NLP retweeted
🎉 Excited to share that our paper “One Form to Transfer Them All” has been accepted to the #EMNLP2026 Main Conference!
We study how pretraining representations shape cross-lingual transfer in multilingual LMs.
🧵
UW NLP retweeted
Super fun to see software world come to life!!
We find codebases with large dependencies and build robust infra & envs where agents can meaningfully collaborate and work together.
Demo: theagentorg.app
UW NLP retweeted
Building an environment to develop and evaluate multi-agent systems at scale is challenging. Software World is our first step towards it:
📍We build a simulated GitHub ecosystem for agents that reflects the realistic needs of experienced software developers.
📍We construct robust, held-out, verifiable downstream applications that are less susceptible to reward hacking.
Check out our blog post to learn more!
theagentorg.app/
UW NLP retweeted
Automatic harness evolution appears to be a promising path toward AI self-improvement, but we find that its gains still largely come from repeated sampling and show limited generalization.
Blog post: yikee.github.io/harnessevolu…
Code: github.com/rethinking-harnes…
UW NLP retweeted
I'm honored to receive the 2025 AAAI/ACM SIGAI Doctoral Dissertation Award!
I recently spoke with SIGAI AI Matters about my dissertation, "Beyond Scaling: Frontiers of Retrieval-Augmented Language Models," and the research journey behind it:
sigai.acm.org/main/ai-matter…
our very own @scottgeng00 on training olmo3 & dpo
New podcast/lecture combo -- a case study in the messy details of Olmo 3 post training & DPO with @scottgeng00. It's rare to make time for these discussions, but we cover:
What it takes for a research idea to make it into a (near) frontier model.
The messy side of DPO (usually data).
Organizational challenges in multi-stage post-training recipes.
Reflections on where research is heading today, and how people should think of DPO.
DPO fundamentals (once more)
Other topics.
Scott is one of the great student's I've had the pleasure of working with for a few years!
Chapters:
00:00 Introduction
04:32 Preference Tuning Basics
08:21 The DPO-Algorithm Zoo
17:06 Scaling Preference Data
21:55 Delta Learning Hypothesis
32:30: Applying it to Olmo 3
36:08 Challenge 1: Data Details
47:17 Challenge 2: Organization Management
51:45 Challenge 3: Scaling Issues
59:15 The Future of Academic Research
Excited to keep making this content under the motivation of my book. Really plugging away now :)
Congrats to @hamishivi @yinn_oscar @RulinShao for their work on tmax 👐
credit to our data chef @yinn_oscar 🧑🍳
cool finding from @TengX6 and @yikewang_ on the evaluation of automatic harness evolution!
Automatic harness evolution appears to be a promising path toward AI self-improvement, but we find that its gains still largely come from repeated sampling and show limited generalization.
Blog post: yikee.github.io/harnessevolu…
Code: github.com/rethinking-harnes…
UW NLP retweeted
Check out this work led by @MA_Shafiei98!
Models act fairer when the prompt looks like a fairness test⚠️
We call this **performative compliance** and introduce a method to reveal morally unsafe behaviors through verifiable demographic puzzles🧩
UW NLP retweeted
Most people think machine translation is solved, but there are so many other challenges even with frontier LLMs.
Multilingual reasoning cascades need more context⚠️
This is the first paper I’ve mentored undergrads on and it’s both an exciting and scary process. I really learned a lot along the process, and so happy that I’m back to working on MT!! Truly a full circle moment when the paper came out🫶
