works on

From the 1 of 10 linked papers with an AI index.

collaborators

10 papers

cs.CL2026

Practical Online KV Cache Compaction for LLM Agents: An Empirical Study

Yujian Liu, Jiabao Ji, Li An +4

LLM agents accumulate long trajectories of reasoning steps, tool calls, and environment feedback, making the KV cache a major inference bottleneck. KV cache compaction can reduce t…

cs.AI2026

Speculate While You Reason: Teaching Agents to Predict Their Next Tool Call via Joint Agent-Speculator RL

Jiabao Ji, Yujian Liu, Li An +4

The paper introduces a self‑speculating agent that unifies an LLM agent and a tool‑call speculator in a single model, using joint reinforcement learning to predict its next tool ca…

cs.CV2026

BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation

Baoyou Chen, Hanchen Xia, Peng Tu +5

Autoregressive vision-language models (VLMs) deliver strong multimodal capability, but their token-by-token decoding imposes a fundamental inference bottleneck. Diffusion VLMs offe…

cs.RO2026

UniTacVLA: Unified Tactile Understanding and Prediction in Vision Language Action Models

Xidong Zhang, Yichi Zhang, Jiaxin Shi +5

Vision-language-action (VLA) models have achieved strong performance in many robotic manipulation tasks, yet remain limited in contact-rich dexterous manipulation. To overcome this…

cs.CV2026

3DThinkVLA: Endowing Vision-Language-Action Models with Latent 3D Priors via 3D-Thinking-Guided Co-training

Jiaxin Shi, Xidong Zhang, Fucai Zhu +3

We propose a 3D-thinking-guided co-training framework that enables vision-language-action (VLA) models to perform 3D spatial reasoning implicitly during action prediction. Our core…

cs.CL2026

T: Progressive Block Scaling for Masked Diffusion Language Models Through Trajectory Aware Reinforcement Learning

Hanchen Xia, Baoyou Chen, Yutang Ge +2

We present T, a simple TraceRL-based training curriculum for progressive block-size scaling in masked diffusion language models (MDMs). Starting from an AR-initialized smal…