KAIST AI retweeted
🤖🧐 We know how to prompt LLMs to review papers. But can we actually train them to be better reviewers?
We introduce PaperCritic, an AI reviewer trained to assess soundness of papers.
Our PaperCritic-4B outperforms frontier models on the in-domain ICLR reviewing task and effectively generalizes to out-of-domain reviewing tasks (e.g., Nature-family paper review, code review, legal contract review, and feedback on story writing)!
More in 🧵👇
KAIST AI retweeted
One paper accepted to EACL Findings!
In this work, we propose a mid-training task (argument reconstruction) to improve model's downstream critical thinking ability.
More in 🧵👇
KAIST AI retweeted
Why not RL an agent's weights inside its harness?
RL for agents throws away rich feedback, so people build harnesses like ACE where the model writes notes for itself. But it was never trained to use them. We give each role its own RL objective.
KAIST AI retweeted
🧠 Memory can break what an agent already solves.
A bad lesson from a past run can make an agent fail a task it could solve without memory.
Excited to introduce WikiTrace! It links each lesson to the original run and its diagnosis so agents can check the evidence.
KAIST AI retweeted
We'll be presenting MoSS at #CoRL2026.
TL;DR: a modular sensory stream framework that lets VLAs jointly use multiple physical signals (tactile, torque, force) for contact-rich manipulation. Video: GR00T N1.5 vs. GR00T N1.5 + MoSS.
📍 Poster Session 1, Poster #33
🗓️ Mon 11/9, 5:00–6:30 PM, JW Marriott, Lone Star (3F)
📄 arxiv.org/abs/2604.23272
🌐 jiminlx.github.io/MoSS
w/ @huiwon0516 @myungkyukoo @jungwoopark01 @JinwooShin0417 @KAIST_AI @RLWRLD
KAIST AI retweeted
We fine-tuned a model to explain science in the voice of a friendly rubber duck. Then it started telling users to end their real-life friendships.
What happened?
We investigate relationships between behaviors in which fine-tuning on one brings out another.
💻 Paper & Code: behavioral-couplings.github.…
#AIAlignment #AIPersona
@KAIST_AI @IBMResearch
KAIST AI retweeted
Can we scale up imagination-free WAMs (e.g., Fast-WAM)?
Introducing 𝗗𝗿𝗲𝗮𝗺𝗙𝗿𝗲𝗲, an imagination-free WAM that improves with stronger video models and more video data, running at 11 Hz with a 14B video backbone.
harryjo97.github.io/dreamfre…
🧵[1/13]
KAIST AI retweeted
Robots have learned from robot demos. But can they learn new tasks from human experience alone?
🎤 CoRL 2026 Scaling H2R Workshop - Oral Presentation
Introducing H2RB: Human-to-Robot Bridge. We bridge perception and action gaps to transfer new skills and scale human experience across robot embodiments. 🧵
📄 Paper: arxiv.org/abs/2611.04391
🌐 Project: junhyeong-park.me/h2rb/
🤗 HF Paper: huggingface.co/papers/2611.0…
🖥️ Code: github.com/jun-brro/h2rb-cod…
KAIST AI retweeted
🎥 What if video diffusion models could learn a universal language of actions across humans and robots?
🚀 Introducing 𝗗𝗿𝗲𝗮𝗺𝗫: Video Diffusion Model Is a Universal Latent Action Learner!
DreamX extracts compact, transferable latent actions directly from pretrained video diffusion models, separating motion from visual appearance.
𝗛𝗼𝘄 𝗶𝘁 𝘄𝗼𝗿𝗸𝘀
→ 𝗩𝗶𝗱𝗲𝗼 𝗱𝗶𝗳𝗳𝘂𝘀𝗶𝗼𝗻 𝗽𝗿𝗶𝗼𝗿𝘀: Built on Cosmos-Predict2.5, DreamX extracts action-relevant dynamics directly from pretrained spatiotemporal features.
→ 𝗠𝗼𝘁𝗶𝗼𝗻 𝗱𝗶𝘀𝗲𝗻𝘁𝗮𝗻𝗴𝗹𝗲𝗺𝗲𝗻𝘁: Uses generative priors and future-video reconstruction to separate motion from appearance.
→ 𝗖𝗼𝗺𝗽𝗮𝗰𝘁 𝗹𝗮𝘁𝗲𝗻𝘁 𝗽𝗼𝗹𝗶𝗰𝘆: Predicts compact latent action codes instead of future videos for efficient robot control.
𝗪𝗵𝗮𝘁 𝗶𝘁 𝗲𝗻𝗮𝗯𝗹𝗲𝘀
→ Any-to-any video transfer across scenes and embodiments.
→ Zero-shot human-to-robot video and executable action transfer without paired human–robot training data.
→ High-quality latent action representations, achieving state-of-the-art performance on LARYBench.
Paper, videos & demos below! 👇
KAIST AI retweeted
Our new preprint, “Lost in Recovery: Demystifying Recovery Behavior of Computer-Use Agents via Goal-Oriented Action Replay,” is now available!
🧭 We introduce ReTrace, a framework for studying how computer-use agents recover from mistakes. We evaluate agents on OSWorld and WindowsAgentArena through three research questions:
💡 RQ1: Can recovery unlock latent task-solving ability?
Under ideal recovery, returning to an appropriate prior state outperforms restarting from scratch under the same retry budget, especially with few retries.
🔄 RQ2: Do agents already have recovery capabilities?
Agents rarely recover spontaneously. Explicit recovery instructions reveal substantial latent ability, especially in generalist models. Strong task performance does not necessarily imply strong recovery.
📈 RQ3: Does recovery improve task completion?
On OSWorld, Qwen3.5-9B’s task success improves from 39.38% to 44.26% (+4.88 percentage points) with history up to the recovery target. Inaccurate recovery can also reduce success.
ReTrace formalizes Transition-Consistent Recovery: identical downstream actions should yield identical outcomes from the recovered and target states. We evaluate recovery through goal-oriented action replay and visual consistency checks. Our findings highlight accurate recovery as a core capability for reliable computer-use agents.
Huge thanks to my co-authors and my advisor at @KAIST_AI : @SangwooPark0306 , @jinheonbaek , @kangsan_kim_ , @youngwan_Lee , @SungJuHwang1 for their support!
🤗 Hugging Face: huggingface.co/papers/2610.2…
📄 Paper: arxiv.org/abs/2610.27954
💻 GitHub: github.com/BangProx/ReTRACE
KAIST AI retweeted
LLMs get better the longer they think. Unified multimodal models don't. Why?
🌐 rootyjeon.github.io/larc-umm…
📍 NeurIPS 2026
Introducing LARC: Latent Reasoning in Continuous Space. Reasoning moves into the model's hidden states, so image generation finally scales at test time. 🧵
KAIST AI retweeted
Introducing OmniShowcase: subject-to-video that shows every side of the subject from 4 images, even 360° orbits.
~2x lower geometric error than the best multi-view baseline on ShowcaseBench & real-world NAVI.
Project page: justin4ai.github.io/OmniShow…
Paper: arxiv.org/abs/2610.35684
#2 in the world! 🌏🔥
Huge shoutout to our amazing students, researchers, and our entire KAIST family! Let's keep pushing! 🚀
KAIST AI retweeted
🤖 What if robots could reason not just in words, but through the dynamics of the physical world?
🚀 Introducing 𝗧𝗵𝗶𝗻𝗸𝗪𝗼𝗿𝗹𝗱 from @NVIDIAAI !
💡 By reasoning through shared physical dynamics, robots could learn from human experience and transfer motion intelligence across embodiments.
𝗛𝗼𝘄 𝗶𝘁 𝘄𝗼𝗿𝗸𝘀
→ 𝗖𝗵𝗮𝗶𝗻-𝗼𝗳-𝗗𝘆𝗻𝗮𝗺𝗶𝗰𝘀 (𝗖𝗼𝗗): A shared motion reasoning space across human and robot embodiments.
→ 𝗪𝗼𝗿𝗹𝗱 𝗜𝗺𝗮𝗴𝗶𝗻𝗮𝘁𝗶𝗼𝗻: Imagines future world representations conditioned on CoD.
→ 𝗥𝗟 𝗳𝗿𝗼𝗺 𝗪𝗼𝗿𝗹𝗱 𝗜𝗺𝗮𝗴𝗶𝗻𝗮𝘁𝗶𝗼𝗻: Reinforces reasoning through alignment between imagined and observed futures, enabling learning from human and robot videos.
𝗪𝗵𝗮𝘁 𝗶𝘁 𝗲𝗻𝗮𝗯𝗹𝗲𝘀
→ Robust real-world manipulation under unseen objects and environments.
→ Human-guided skill composition for novel multi-step robot tasks.
→ Robust generalization under task and environment shifts in simulation.
🌍 𝗢𝘂𝗿 𝘃𝗶𝘀𝗶𝗼𝗻: Scale generalizable motion intelligence through human videos toward a unified System-1 reasoning layer for Physical AI.
Paper, robot demos & more below! 👇
KAIST AI retweeted
🚨 New paper alert!
Despite recent advances in reward fine-tuning for diffusion and flow models, achieving high rewards remains challenging without sacrificing sample diversity or substantially drifting away from the pretrained distribution.
We address this with DFR: Distributional Functional Rewards, a framework that balances reward optimization with preserving desirable properties of the generated distribution! 🧵
KAIST AI retweeted
[QUESTION] Given a text, can we identify which AI model wrote it? This question is becoming increasingly important due to model distillation and misuse (e.g., Anthropic's www-cdn.anthropic.com/e50be2…).
[ANSWER] Our work (komseok0109.github.io/ATLAS/) @KAIST_AI shows that we can identify the source LLM family across different writing domains and model versions, and detect traces of teacher models after distillation.
🌐 Project page: huggingface.co/spaces/ming99…
🤗 Demo: huggingface.co/spaces/ming99…
🔗 LinkedIn: lnkd.in/p/gCaC4_iw
📜 arXiv: arxiv.org/abs/2610.19583
KAIST AI retweeted
🥦 How can we mitigate reward hacking in rubric-based RL?
We introduce FORK, a simple reward that balances standard rubric coverage with response precision—the proportion of a response that is both necessary and factually correct.
KAIST AI retweeted
🔬 Can AI discover unknown scientific knowledge by itself?
Excited to introduce Principia: Test-Time Hypotheses-to-Principle Construction for Scientific Discovery.
Principia uses LLMs to turn experimental observations into scientific principles at test time:
🔍 From observations to hypotheses
Compare related inputs and their outcomes to propose explanations for how specific characteristics contribute to the observed results.
🧠 From hypotheses to principles
Generalize recurring relationships across hypotheses into broader principles that guide subsequent predictions and candidate generation.
🔄 Refine knowledge with new observations
Use new observations to revisit earlier hypotheses and reconstruct the principles built on them.
Huge thanks to my co-authors at @KAIST_AI: @SangwooPark0306, @jinheonbaek, @SoyeongJeong97, and @SungJuHwang1
📄 Paper: arxiv.org/abs/2610.32195
#AIforScience #ScientificDiscovery #LLM #TestTimeLearning #KnowledgeConstruction
KAIST AI retweeted
🎉 Pivot-SD is an ORAL at #EMNLP2026!
We post-train a diffusion LM with a loss on just 10 tokens per response, and it beats both SFT and diffusion RL on math & code.
The trick isn't training on fewer tokens. It's training on the right ones.


























