Chinese Academic of Science Institute of Automation

university

Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

JokerJan submitted a paper about 16 hours ago

Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning

JokerJan submitted a paper about 16 hours ago

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

zd21 submitted a paper 4 days ago

Continual Learning in Transition

View all activity

Papers

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

Continual Learning in Transition

View all Papers

CASIA 's papers 22

Submitted by

zkj

SFT Conflicts, RL Coexists: A Theoretical and Empirical Analysis of Multi-Task Learning for LLMs

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Dan Zhang

Continual Learning in Transition

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

zkj

Beyond Simply Environment Scaling: Designing Effective Environment Distributions for Multimodal Agent Learning

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

TanYuQiao

SWE-Touch: Benchmarking Coding Agents When Users Touch the Code

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

taesiri

PhiZero: A World Model Built Around Physical Language

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Zhuoran Jin

The Physics of Multi-Turn Long-Horizon Planning: From Pre-training to Post-training via Single- and Multi-Teacher On-Policy Agentic Distillation

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Zhuoran Jin

Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Zhuoran Jin

Look Light, Think Heavy: What Multimodal Chain-of-Thought Reasoning Can and Cannot Do

CASIA

Chinese Academic of Science Institute of Automation

1

Submitted by

Hongbo Wang

Coloring the Noise: Adversarial Sobolev Alignment for Faithful Image Super Resolution

CASIA

Chinese Academic of Science Institute of Automation

Submitted by

Xiaotian Ye

Uncovering Entity Identity Confusion in Multimodal Knowledge Editing

CASIA

Chinese Academic of Science Institute of Automation

2

Submitted by

zihan

ResRL: Boosting LLM Reasoning via Negative Sample Projection Residual Reinforcement Learning

CASIA

Chinese Academic of Science Institute of Automation

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs

CASIA

Chinese Academic of Science Institute of Automation