13 papers
CoT-Core: Accelerating LLM Evaluation via CoT-Aware Coreset Selection
Qihua Pan, Zhenheng Tang, Peijie Dong +4
Evaluating Large Language Models (LLMs) incurs prohibitive computational overhead during continuous development processes. While coreset selection accelerates evaluation, existing…
Position: LLM Inference Should Be Evaluated as Energy-to-Token Production
Xiang Liu, Shimiao Yuan, Zhenheng Tang +5
LLM inference is still evaluated mainly as a model or software problem: accuracy, latency, throughput, and hardware utilization. This is incomplete. At deployment scale, the releva…
Dual-Dimensional Consistency: Balancing Budget and Quality in Adaptive Inference-Time Scaling
Rongman Xu, Yifei Li, Tianzhe Zhao +3
Large Language Models (LLMs) have demonstrated remarkable abilities in reasoning. However, maximizing their potential through inference-time scaling faces challenges in trade-off b…
Is Your LLM-as-a-Recommender Agent Trustable? LLMs' Recommendation is Easily Hacked by Biases (Preferences)
Zichen Tang, Zirui Zhang, Qian Wang +3
Current Large Language Models (LLMs) are gradually exploited in practically valuable agentic workflows such as Deep Research, E-commerce recommendation, and job recruitment. In the…
Are Dilemmas and Conflicts in LLM Alignment Solvable? A View from Priority Graph
Zhenheng Tang, Xiang Liu, Qian Wang +3
As Large Language Models (LLMs) become more powerful and autonomous, they increasingly face conflicts and dilemmas in many scenarios. We first summarize and taxonomize these divers…
On the Spectral Flattening of Quantized Embeddings
Junlin Huang, Wenyi Fang, Zhenheng Tang +5
Training Large Language Models (LLMs) at ultra-low precision is critically impeded by instability rooted in the conflict between discrete quantization constraints and the intrinsic…
RouteMark: A Fingerprint for Intellectual Property Attribution in Routing-based Model Merging
Xin He, Junxi Shen, Zhenheng Tang +4
Model merging via Mixture-of-Experts (MoE) has emerged as a scalable solution for consolidating multiple task-specific models into a unified sparse architecture, where each expert…
RA-NeRF: Robust Neural Radiance Field Reconstruction with Accurate Camera Pose Estimation under Complex Trajectories
Qingsong Yan, Qiang Wang, Kaiyong Zhao +4
Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have emerged as powerful tools for 3D reconstruction and SLAM tasks. However, their performance depends heavily on ac…
Can Compressed LLMs Truly Act? An Empirical Evaluation of Agentic Capabilities in LLM Compression
Peijie Dong, Zhenheng Tang, Xiang Liu +3
Post-training compression reduces the computational and memory costs of large language models (LLMs), enabling resource-efficient deployment. However, existing compression benchmar…
The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve?
Zhenheng Tang, Xiang Liu, Qian Wang +4
Motivated by reducing the computational and storage costs of LLMs, model compression and KV cache compression have attracted much attention from researchers. However, current metho…
Mediator: Memory-efficient LLM Merging with Less Parameter Conflicts and Uncertainty Based Routing
Kunfeng Lai, Zhenheng Tang, Xinglin Pan +7
Model merging aggregates Large Language Models (LLMs) finetuned on different tasks into a stronger one. However, parameter conflicts between models leads to performance degradation…
Semantic Integrity Matters: Benchmarking and Preserving High-Density Reasoning in KV Cache Compression
Xiang Liu, Zhenheng Tang, Hong Chen +6
While Key-Value (KV) cache compression is essential for efficient LLM inference, current evaluations disproportionately focus on sparse retrieval tasks, potentially masking the deg…
ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
Xiang Liu, Zhenheng Tang, Peijie Dong +5
Large Language Models (LLMs) require significant GPU memory when processing long texts, with the key value (KV) cache consuming up to 70\% of total memory during inference. Althoug…