phd-ing @csatusc on theoretical foundations of LLMs | alum @uchicago, B.S. '20 | student researcher @googleresearch | prev research intern @metaai
Los Angeles
Joined November 2020
- Tweets350
- Following1K
- Followers1.1K
- Likes1K
Deqing Fu retweeted
Iโll be presenting this work at #EMNLP2026 on
๐๏ธMonday, Oct. 26 in ๐Demo Session E, Hall H at 4:45pm!
Also, Iโm excited to chat about human-AI collaboration, societal impacts of AI, and safety risks of language models in mental health interactions. Ping me here or on Whova and letโs connect over coffee โ๏ธ !
๐ก How can experts turn unstructured text into useful insights without weeks of writing annotation guidelines?
Meet CoDA: a human-LM collaborative tool that develops codebooks in <2 hours, using 5ร fewer examples than manual codebook development.
๐coda.44.218.57.167.sslip.io/
๐ arxiv.org/abs/2610.24157
1/10
Deqing Fu retweeted
I'll be at #EMNLP2026 in Budapest ๐ญ๐บ presenting MMGR at Poster Session A in Hall H on October 25, 11:00โ12:30!
MMGR asks a simple but important question: do generative models actually reason ๐ก, or do they just produce outputs that look plausible ๐?
We introduce a unified benchmark for evaluating reasoning in image and video generation across 10 tasks spanning abstract reasoning ๐งฎ, embodied navigation ๐ค, and physical commonsense ๐๐ณ with By evaluating physical, logical, spatial, and temporal reasoning, we find that visual quality doesn't necessarily translate into reasoning correctness.
Read more about MMGR: arxiv.org/abs/2512.14691
Huge shout-out to my AMAZING coauthors @Zefan_Cai @HaoyiQiu @TianyiM15933, Haozhe Zhao, and the rest of the team ๐ฅฐ
Happy to chat about multimodal generative reasoning, agentic search, synthetic data curation and human-AI collaborations! I am always excited to catch up with old friends, meet new friends, exchange ideas, and of course, explore Budapest! ๐ญ๐บ If you're around at #EMNLP2026, come say hi! ๐
Expo ๐ Wilshire seems like a direct route from USC to Sun Nong Dan ๐
LA Metro broke ground this week on two significant transit projects: The Vermont Avenue BRT (NET 2028) and the East San Fernando Valley (NET 2031). The Wilshire subway (D) line will be wrapping up over the next couple years, NoHo-Pas BRT, and Airtrain same. These projects will add 63 new rapid transit stops to our network.
Deqing Fu retweeted
Iโll be at NeurIPS 2026 in Atlanta, Dec 9โ13! Looking forward to catching up with friends, meeting new people, and chatting about exciting ideas. Hope to see you there! ๐
Excited to share Economy of Minds (EoM), where simple local incentives give rise to emergent coordination, specialization, and self-improvement โ no central manager needed! ๐ฐ๐ค
Weโll be presenting this work at NeurIPS 2026 #NeurIPS2026 at the Atlanta ๐ง site! ๐ Come chat with us about Economy of Minds, multi-agent systems, emergent intelligence, and what it takes to build self-organizing societies of AI agents!
Deqing Fu retweeted
Introducing ๐๐จ๐๐๐, my summer work at @GoogleDeepMind
๐๐ฎ๐ป ๐๐๐ ๐ ๐น๐ฒ๐ฎ๐ฟ๐ป ๐๐ผ ๐ณ๐ฎ๐ฐ๐ถ๐น๐ถ๐๐ฎ๐๐ฒ ๐ต๐๐บ๐ฎ๐ป ๐ด๐ฟ๐ผ๐๐ฝ ๐ฐ๐ผ๐ป๐๐ฒ๐ฟ๐๐ฎ๐๐ถ๐ผ๐ป๐ ๐ฎ๐ฐ๐ฟ๐ผ๐๐ ๐ฑ๐ถ๐๐ฒ๐ฟ๐๐ฒ ๐ฑ๐ผ๐บ๐ฎ๐ถ๐ป๐โ๐๐ถ๐๐ต๐ผ๐๐ ๐บ๐ฎ๐ป๐๐ฎ๐น๐น๐ ๐ฑ๐ฒ๐๐ถ๐ด๐ป๐ฒ๐ฑ ๐๐๐ฟ๐ฎ๐๐ฒ๐ด๐ถ๐ฒ๐? ๐ค
๐ก ๐ง๐ต๐ฒ ๐๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ
Good facilitation goes beyond task success. It should also improve ๐ต๐๐บ๐ฎ๐ป ๐ฒ๐
๐ฝ๐ฒ๐ฟ๐ถ๐ฒ๐ป๐ฐ๐ฒ and promote ๐ณ๐ฎ๐ถ๐ฟ๐ป๐ฒ๐๐, ensuring that everyone feels heard, comfortable, and respected.
We study three settings: open-ended discussions, consensus-building, and negotiation.
๐ง ๐ข๐๐ฟ ๐๐ฝ๐ฝ๐ฟ๐ผ๐ฎ๐ฐ๐ต
Rather than manually engineering task-specific strategies, GUIDE learns a ๐ด๐ฒ๐ป๐ฒ๐ฟ๐ฎ๐น๐ถ๐๐ฎ๐ฏ๐น๐ฒ ๐ณ๐ฎ๐ฐ๐ถ๐น๐ถ๐๐ฎ๐๐ถ๐ผ๐ป ๐ฝ๐น๐ฎ๐๐ฏ๐ผ๐ผ๐ธ from 432 real human conversations, iteratively refining it through reflection and human-aligned LLM evaluation.
๐ ๐๐ฒ๐ ๐๐ถ๐ป๐ฑ๐ถ๐ป๐ด๐
๐ ๐ฒ๐ญ.๐ฐ% offline win rate against task-specific expert facilitators.
๐ค Across ๐ฐ๐ฒ๐ฎ live conversations, GUIDE improves speaking comfort, reduces interpersonal friction, and promotes more equitable experiences, while maintaining comparable task outcomes.
๐ฌ ๐๐ฒ๐๐ ๐ถ๐ ๐บ๐ผ๐ฟ๐ฒ: ๐ฎ๐ฐ.๐ณ% fewer interventions and ๐ฎ๐ต.๐ฌ% fewer words!
Huge thanks to my amazing collaborators and mentors @tzushengkuo @Jiale_Leo @DeqingFu @Nithum and Crystal Qian!
๐ Paper: arxiv.org/abs/2610.16231
#GoogleDeepMind #LLM #HumanAIInteraction #SocialIntelligence #AIResearch
TimeRSI: a coding agent wraps a frozen forecasting model, turns domain knowledge + data insights into executable pipelines, and uses online forecast feedback to keep improving as conditions change.
The TSFM stays frozen. The deployment makes it smarter.
Same lesson for tabular and time series: keep the FM frozen, let an agent own the deployment.
It's time for the second half of TSFMs: agentic deployment that adapts them to the real world, rather than expecting frozen models to handle everything alone.
๐ Paper: arxiv.org/abs/2610.19725
A more efficient 2.0 is on the way.
This is the work lead mainly by Haoxin Liu (linkedin.com/in/haoxin-liu-0โฆ)!
TabFM-Auto (xitter.cf/DeqingFu/status/210518โฆ) showed that a coding agent evolving the pipeline around a frozen tabular FM can beat the field.
Time-series foundation models need a harness, too. ๐
Introducing TimeRSI: a self-improving agentic deployment system for TSFMs ๐งต
With TimeRSI, TimesFM-3.0 returns to #1 on TIME๐ and FEV-bench๐ as of Oct15.
Introducing TabFM-Auto, my summer project at @GoogleResearch .
LLMs possess rich world knowledge yet bad at handling massive tables with millions of number cells.
Tabular Foundation Models are so good at tabular in-context learning but don't have semantic knowledges.
Deqing Fu retweeted
Excited to share PreciseCoder, a natural next step from our Precise Debugging Benchmark. PDB showed that coding models can pass tests while making many unnecessary edits. This time, we asked: can we train them to fix the bug and leave working code alone?
precise-coder-rl.github.io/
We found a tension in the usual training recipe.
(1) Unit-test rewards care whether the final program passes, regardless of what changed.
(2) Simply penalizing edit size tends to shrink patches at the cost of pass rate.
PreciseCoder instead pairs test success with a reward that verifies proposed edits against reference faults. On PDB, our 4B model cuts edits from 14 to 3 lines per task at the same pass rate. The gains carry over to harder, repository-level and agentic debugging.
The finding Iโm most excited about is weak-to-strong transfer: give the reasoning trace of our 9B model to a stronger model, and its edit precision improves by 4.6โ12.4 points without further training. Our analysis of those reasoning traces offers a clue: PreciseCoder more often names the buggy line and reasons about whether a change is needed.
Proud of @zhanyunkai and the whole team for this work! โ๏ธ
Stop letting your coding agents BLOW UP your codebase. ๐ฅ
Only 62% of the lines GPT-5.6 Sol edits are actually needed when debugging on PDB;
DeepSeek-V4.1-Flash changes 60% of already-correct code and breaks 22% of it.
A big reason is RL with unit-test rewards; it checks whether the code passes, never what you changed.
Introducing our new paper, PreciseCoder: Faithful and Generalizable Debugging Beyond Unit-Test Rewards
โ๏ธ 14 โ 3 edited lines per task at the same pass rate
๐ก๏ธ leaves correct code untouched 2x as often
๐ generalizes to complex debugging and agentic coding tasks
๐ง teaches frontier models to edit more precisely
precise-coder-rl.github.io
(1/11)
Tired of coding agents rewriting your entire codebase every time you ask them to fix one tiny bug? check out PreciseCoder at precise-coder-rl.github.io/
Stop letting your coding agents BLOW UP your codebase. ๐ฅ
Only 62% of the lines GPT-5.6 Sol edits are actually needed when debugging on PDB;
DeepSeek-V4.1-Flash changes 60% of already-correct code and breaks 22% of it.
A big reason is RL with unit-test rewards; it checks whether the code passes, never what you changed.
Introducing our new paper, PreciseCoder: Faithful and Generalizable Debugging Beyond Unit-Test Rewards
โ๏ธ 14 โ 3 edited lines per task at the same pass rate
๐ก๏ธ leaves correct code untouched 2x as often
๐ generalizes to complex debugging and agentic coding tasks
๐ง teaches frontier models to edit more precisely
precise-coder-rl.github.io
(1/11)
Deqing Fu retweeted
A very overdue life update: came across my graduation photos and realized I never shared them!
I finished my PhD at @CSatUSC ! ๐ So grateful to my advisors @robinomial and @_jessethomason_ for all their guidance and support, and to my partner for being there through all the ups and downs. โค๏ธ
Iโve since joined @UCBerkeley as a postdoc, working with @serinachang5! ๐ป
Deqing Fu retweeted
Can a unified multimodal model (UMM) teach itself, with its text and visual abilities making training data for each other?
Yes: over 4 rounds, a 7B model climbs 45.7% โ 60.2%, vs. 46.3% for continued training.
๐งต 1/10
Deqing Fu retweeted
What's the role of small-scale research in the era of massive models? Tell us where you stand and help decide what the @MOSS_workshop panel debates at #COLM2026 by voting on the poll below! ๐ฑ
Help us shape the panel at MOSS @COLM_conf !
We've put together a quick poll. Questions with the most split opinions will become panel discussion topics.
Vote: form.typeform.com/to/UOyzADcโฆ
See more info on our website: sites.google.com/view/moss-cโฆ
Come and talk with Harvey at COLM! If you know any spring/summer opportunities, please DM him as well!!!
Iโll be at #COLM2026 next week, presenting our work on effective KV-cache eviction at the Efficient Reasoning Workshop on Oct 9, 12:30โ13:30.
I work on LLM capabilities, efficient reasoning, and evaluation, and would be happy to chat with anyone working on related problems!
Iโm also looking for Spring/Summer 2027 internships. Feel free to DM me if you think I'd be a good fit.
Deqing Fu retweeted
Iโll be at #COLM2026 next week, presenting our work on effective KV-cache eviction at the Efficient Reasoning Workshop on Oct 9, 12:30โ13:30.
I work on LLM capabilities, efficient reasoning, and evaluation, and would be happy to chat with anyone working on related problems!
Iโm also looking for Spring/Summer 2027 internships. Feel free to DM me if you think I'd be a good fit.
