22 papers
SafeRI: Recognition and Intervention for Token-Level Safety Intervention in Large Vision Language Models
Caoyuan Ma, Tian Gu, Wenpu Liu +11
Existing safety alignment methods for vision-language models usually modify the model behavior globally: once the safety parameters are trained or loaded, they participate in both…
OptiMAS: Automatically Optimize Multi-Agent System
Yuxin Cheng, Chang Liu, Hanxin Yu +9
Automated evolution of Multi-Agent Systems (MAS) holds significant potential for reducing the manual effort required to design and optimize LLM-based agent architectures. However,…
PatchINR: Patch-Based Implicit Neural Representations for Efficient and Scalable Inference
Jiachen Ren, Wenyong Zhou, Taiqiang Wu +4
Implicit Neural Representation (INR) provides an effective approach for continuous signal modeling, but classical per-pixel inference results in quadratic growth in inference count…
PrefixFlow: Training-Time KV Caching for Schedule-Level Prefix Reuse in LLM RL Training
Pengbo Li, Feiyuan Zhang, Guangming Sheng +8
In reinforcement learning (RL) post-training, one prompt fans out into trajectories, yet standard trainers replay the shared prefix's forward and backward pass times; in lo…
What Makes Interaction Trajectories Effective for Training Terminal Agents?
Sidi Yang, Chaofan Tao, Jierun Chen +11
Stronger code agents are commonly assumed to be superior teachers for post-training, yet this assumption remains poorly disentangled from task difficulty, harness design, and stude…
ROMER: Expert Replacement and Router Calibration for Robust MoE LLMs on Analog Compute-in-Memory Systems
Wenyong Zhou, Yuannuo Feng, Yizhe Chen +6
Large language models (LLMs) with mixture-of-experts (MoE) architectures achieve remarkable scalability by sparsely activating a subset of experts per token, yet their frequent exp…
Attention Sink in Transformers: A Survey on Utilization, Interpretation, and Mitigation
Zunhai Su, Hengyuan Zhang, Wei Wu +24
As the foundational architecture of modern machine learning, Transformers have driven remarkable progress across diverse AI domains. Despite their transformative impact, a persiste…
The Art of Efficient Reasoning: Data, Reward, and Optimization
Taiqiang Wu, Zenan Xu, Bo Zhou +1
Large Language Models (LLMs) consistently benefit from scaled Chain-of-Thought (CoT) reasoning, but also suffer from heavy computational overhead. To address this issue, efficient…
Can We Trust LLMs on Memristors? Diving into Reasoning Ability under Non-Ideality
Taiqiang Wu, Yuxin Cheng, Chenchen Ding +5
Memristor-based analog compute-in-memory (CIM) architectures provide a promising substrate for the efficient deployment of Large Language Models (LLMs), owing to superior energy ef…
BPDQ: Bit-Plane Decomposition Quantization on a Variable Grid for Large Language Models
Junyu Chen, Jungang Li, Jing Xiong +11
Large language model inference is often bounded by memory footprint and bandwidth in resource-constrained deployments, making quantization fundamental to efficient serving. While p…
MMFormalizer: Multimodal Autoformalization in the Wild
Jing Xiong, Qi Han, Yunta Hsieh +11
Autoformalization, which translates natural language mathematics into formal statements to enable machine reasoning, faces fundamental challenges in the wild due to the multimodal…
Re-Activating Frozen Primitives for 3D Gaussian Splatting
Yuxin Cheng, Binxiao Huang, Wenyong Zhou +4
3D Gaussian Splatting (3D-GS) achieves real-time photorealistic novel view synthesis, yet struggles with complex scenes due to over-reconstruction artifacts, manifesting as local b…
Perspective-aware 3D Gaussian Inpainting with Multi-view Consistency
Yuxin Cheng, Binxiao Huang, Taiqiang Wu +5
3D Gaussian inpainting, a critical technique for numerous applications in virtual reality and multimedia, has made significant progress with pretrained diffusion models. However, e…
Revisiting Model Interpolation for Efficient Reasoning
Taiqiang Wu, Runming Yang, Tao Liu +2
Model merging, typically on Instruct and Thinking models, has shown remarkable performance for efficient reasoning. In this paper, we systematically revisit the simplest merging me…
Timber: Training-free Instruct Model Refining with Base via Effective Rank
Taiqiang Wu, Runming Yang, Tao Liu +3
Post-training, which elicits a pretrained Base model into the corresponding Instruct model, is widely considered to be superficial. In this work, we first reinforce this hypothesis…
QuadINR: Hardware-Efficient Implicit Neural Representations Through Quadratic Activation
Wenyong Zhou, Boyu Li, Jiachen Ren +4
Implicit Neural Representations (INRs) encode discrete signals continuously while addressing spectral bias through activation functions (AFs). Previous approaches mitigate this bia…
Enhancing Robustness of Implicit Neural Representations Against Weight Perturbations
Wenyong Zhou, Yuxin Cheng, Zhengwu Liu +3
Implicit Neural Representations (INRs) encode discrete signals in a continuous manner using neural networks, demonstrating significant value across various multimedia applications.…
Distribution-Aware Hadamard Quantization for Hardware-Efficient Implicit Neural Representations
Wenyong Zhou, Jiachen Ren, Taiqiang Wu +3
Implicit Neural Representations (INRs) encode discrete signals using Multi-Layer Perceptrons (MLPs) with complex activation functions. While INRs achieve superior performance, they…
MINR: Efficient Implicit Neural Representations for Multi-Image Encoding
Wenyong Zhou, Taiqiang Wu, Zhengwu Liu +3
Implicit Neural Representations (INRs) aim to parameterize discrete signals through implicit continuous functions. However, formulating each image with a separate neural network~(t…
SwingArena: Competitive Programming Arena for Long-context GitHub Issue Solving
Wendong Xu, Jing Xiong, Chenyang Zhao +16
We present SwingArena, a competitive evaluation framework for Large Language Models (LLMs) that closely mirrors real-world software development workflows. Unlike traditional static…
Shadow-FT: Tuning Instruct Model via Training on Paired Base Model
Taiqiang Wu, Runming Yang, Jiayi Li +4
Large language models (LLMs) consistently benefit from further fine-tuning on various tasks. However, we observe that directly tuning the Instruct (i.e., instruction-tuned) models…
HaLoRA: Hardware-aware Low-Rank Adaptation for Large Language Models Based on Hybrid Compute-in-Memory Architecture
Taiqiang Wu, Chenchen Ding, Wenyong Zhou +7
Low-rank adaptation (LoRA) is a predominant parameter-efficient finetuning method for adapting large language models (LLMs) to downstream tasks. Meanwhile, Compute-in-Memory (CIM)…