16 papers
AvoKV-E: Payload-Aware KV Cache Eviction for Long Reasoning
Han Yu, Wenhui Zhu, Xiwen Chen +9
Long-output reasoning shifts the KV-cache bottleneck from the fixed prompt to the generated trace. Existing reasoning-cache eviction methods largely treat cached entries as routing…
LatentMT: Machine Translation with Latent Reasoning
Wei-Rui Chen, Samar M. Magdy, Chiyu Zhang +3
Latent-reasoning looped language models (LoopLMs) offer a different scaling path for machine translation (MT): instead of increasing parameter count or emitting explicit chain-of-t…
S-SPPO: Semantic-Calibrated Self-Play Preference Optimization
Xiwen Chen, Wenhui Zhu, Jingjing Wang +13
Aligning Large Language Models (LLMs) with human preferences is often formulated via Direct Preference Optimization (DPO). However, the standard Bradley-Terry instantiation of DPO…
Mags-RL: Wearing Multimodal LLMs a Magnifying Glass via Agentic Reinforcement Learning For Complex Scene Reasoning
Xuanzhao Dong, Wenhui Zhu, Peijie Qiu +11
Despite their popularity and success, Multimodal Large Language Models (MLLMs) often struggle to interpret images accurately, which limits their reasoning capability in complex sce…
OphIn-500K: Curating Web-Scale Visual Instructions for Scaling Ophthalmic Multimodal Large Language Models
Xuanzhao Dong, Wenhui Zhu, Xiwen Chen +13
The advancement of general medical Multimodal Large Language Models (MLLMs) has shown great potential for building conversational assistants to support clinical diagnosis. However,…
Your Agent is More Brittle Than You Think: Uncovering Indirect Injection Vulnerabilities in Agentic LLMs
Wenhui Zhu, Xuanzhao Dong, Xiwen Chen +7
The rapid deployment of open-source frameworks has significantly advanced the development of modern multi-agent systems. However, expanded action spaces, including uncontrolled pri…
Bridging Restoration and Diagnosis: A Comprehensive Benchmark for Retinal Fundus Enhancement
Xuanzhao Dong, Wenhui Zhu, Xiwen Chen +8
Over the past decade, generative models have demonstrated success in enhancing fundus images. However, the evaluation of these models remains a challenge. A benchmark for fundus im…
SODA: Semi On-Policy Black-Box Distillation for Large Language Models
Xiwen Chen, Jingjing Wang, Wenhui Zhu +7
Black-box knowledge distillation for large language models presents a strict trade-off. Simple off-policy methods (e.g., sequence-level knowledge distillation) struggle to correct…
OTPrune: Distribution-Aligned Visual Token Pruning via Optimal Transport
Xiwen Chen, Wenhui Zhu, Gen Li +9
Multi-modal large language models (MLLMs) achieve strong visual-language reasoning but suffer from high inference cost due to redundant visual tokens. Recent work explores visual t…
AriadneMem: Threading the Maze of Lifelong Memory for LLM Agents
Wenhui Zhu, Xiwen Chen, Zhipeng Wang +11
Long-horizon LLM agents require memory systems that remain accurate under fixed context budgets. However, existing systems struggle with two persistent challenges in long-term dial…
EZBlender: Efficient 3D Editing with Plan-and-ReAct Agent
Hao Wang, Wenhui Zhu, Shao Tang +8
As a cornerstone of the modern digital economy, 3D modeling and rendering demand substantial resources and manual effort when scene editing is performed in the traditional manner.…
AHA: Aligning Large Audio-Language Models for Reasoning Hallucinations via Counterfactual Hard Negatives
Yanxi Chen, Wenhui Zhu, Xiwen Chen +9
Although Large Audio-Language Models (LALMs) deliver state-of-the-art (SOTA) performance, they frequently suffer from hallucinations, e.g. generating text not grounded in the audio…
VAOT: Vessel-Aware Optimal Transport for Retinal Fundus Enhancement
Xuanzhao Dong, Wenhui Zhu, Yujian Xiong +8
Color fundus photography (CFP) is central to diagnosing and monitoring retinal disease, yet its acquisition variability (e.g., illumination changes) often degrades image quality, w…
Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
Siyu Zhu, Yanbin Jiang, Hejian Sang +6
We investigated Agentic RL with large language models on the \textsc{TravelPlanner} benchmark. Our approach, \textsc{Planner-R1}, achieved a \textbf{56.9\%} final-pass rate with on…
EVTP-IVS: Effective Visual Token Pruning For Unifying Instruction Visual Segmentation In Multi-Modal Large Language Models
Wenhui Zhu, Xiwen Chen, Zhipeng Wang +5
Instructed Visual Segmentation (IVS) tasks require segmenting objects in images or videos based on natural language instructions. While recent multimodal large language models (MLL…
LLaDA-MedV: Exploring Large Language Diffusion Models for Biomedical Image Understanding
Xuanzhao Dong, Wenhui Zhu, Xiwen Chen +5
Autoregressive models (ARMs) have long dominated the landscape of biomedical vision-language models (VLMs). Recently, masked diffusion models such as LLaDA have emerged as promisin…