2 citations · 5 across the 8 of their papers we have counts for
12 papers
SYNLAT: Syntax-Aligned Text-Latent Compression for Chain-of-Thought Reasoning
Yifeng Zhao, Hongjun Yu, Shibo Wang +5
Long chain-of-thought (CoT) traces impose substantial output-token costs. Under constrained budgets, compression must preserve answer-critical information, making boundary placemen…
Parasitic Co-Denoising: Unlocking 3D Human Motion Generation in a Frozen Video Diffusion Model
Yunjiao Zhou, Junlang Qian, Lihua Xie +1
Despite never being supervised on explicit 3D motion, large-scale text-to-video diffusion models synthesize realistic human motion in their generated videos. We ask whether this im…
Rethinking Fixed Temporal Grids: Frequency-Disentangled Motion Generation
Yunjiao Zhou, Junlang Qian, Gen Li +3
Most human motion generation methods encode motion as tokens on a uniform temporal grid, where every token spans the same fixed time window. Human motion, however, is temporally he…
Rewards as Labels: Revisiting RLVR from a Classification Perspective
Zepeng Zhai, Meilin Chen, Jiaxuan Zhao +3
Reinforcement Learning with Verifiable Rewards has recently advanced the capabilities of Large Language Models in complex reasoning tasks by providing explicit rule-based supervisi…
Fine-Grained Activation Steering: Steering Less, Achieving More
Zijian Feng, Tianjiao Li, Zixiao Zhu +7
Activation steering has emerged as a cost-effective paradigm for modifying large language model (LLM) behaviors. Existing methods typically intervene at the block level, steering t…
Zero-Shot Open-Vocabulary Human Motion Grounding with Test-Time Training
Yunjiao Zhou, Xinyan Chen, Junlang Qian +2
Understanding complex human activities demands the ability to decompose motion into fine-grained, semantic-aligned sub-actions. This motion grounding process is crucial for behavio…
Via Score to Performance: Efficient Human-Controllable Long Song Generation with Bar-Level Symbolic Notation
Tongxi Wang, Yang Yu, Qing Wang +1
Song generation is regarded as the most challenging problem in music AIGC; nonetheless, existing approaches have yet to fully overcome four persistent limitations: controllability,…
Beyond the Next Token: Towards Prompt-Robust Zero-Shot Classification via Efficient Multi-Token Prediction
Junlang Qian, Zixiao Zhu, Hanzhang Zhou +3
Zero-shot text classification typically relies on prompt engineering, but the inherent prompt brittleness of large language models undermines its reliability. Minor changes in prom…
Logit Separability-Driven Samples and Multiple Class-Related Words Selection for Advancing In-Context Learning
Zhu Zixiao, Feng Zijian, Zhou Hanzhang +2
Effective organization of in-context learning (ICL) demonstrations is key to improving the quality of large language model (LLM) responses. To create better sample-label pairs that…
Unveiling and Manipulating Prompt Influence in Large Language Models
Zijian Feng, Hanzhang Zhou, Zixiao Zhu +2
Prompts play a crucial role in guiding the responses of Large Language Models (LLMs). However, the intricate role of individual tokens in prompts, known as input saliency, in shapi…
UniBias: Unveiling and Mitigating LLM Bias through Internal Attention and FFN Manipulation
Hanzhang Zhou, Zijian Feng, Zixiao Zhu +2
Large language models (LLMs) have demonstrated impressive capabilities in various tasks using the in-context learning (ICL) paradigm. However, their effectiveness is often compromi…
LLMs Learn Task Heuristics from Demonstrations: A Heuristic-Driven Prompting Strategy for Document-Level Event Argument Extraction
Hanzhang Zhou, Junlang Qian, Zijian Feng +3
In this study, we investigate in-context learning (ICL) in document-level event argument extraction (EAE) to alleviate the dependency on large-scale labeled data for this task. We…