Robust Length Prediction: A Perspective from Heavy-Tailed Prompt-Conditioned Distributions
State Key Laboratory of Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室);School of Artificial Intelligence, Nanjing University(南京大学人工智能学院)
Output-length prediction is important for efficient LLM serving, as it directly affects batching, memory reservation, and scheduling. For prompt-only length prediction, most existing methods use a one-shot sampled length as the label, implicitly treating each prompt as if it had one true target length. We show that this is unreliable: even under a fixed model and decoding setup, the same prompt induces a \emph{prompt-conditioned output length distribution}, not a deterministic scalar, and this distribution is consistent with \emph{heavy-tailed} behavior. Motivated by this, we cast length prediction as robust estimation from heavy-tailed prompt-conditioned length distributions. We propose prompt-conditioned length distribution (ProD) methods, which construct training targets from multiple independent generations of the same prompt. Two variants are developed to reuse the served LLM's hidden states: \mbox{ProD-M}, which uses a median-based target for robust point prediction, and ProD-D, which uses a distributional target that preserves prompt-conditioned uncertainty. We provide theoretical justifications by analyzing the estimation error under a surrogate model. Experiments across diverse scenarios show consistent gains in prediction quality.
LPT++: Efficient Training on Mixture of Long-tailed Experts
School of Computer Science and Technology, Harbin Institute of Technology(哈尔滨工业大学计算机科学与技术学院);School of Computing and Information Systems, Singapore Management University(新加坡管理大学计算与信息系统学院);The Hong Kong Polytechnic University(香港理工大学)
Decoupled and Divergence-Conditioned Prompt for Multi-domain Dynamic Graph Foundation Models
School of Computer Science and Engineering, Beihang University(北航计算机科学与工程学院);Key Lab of Education Blockchain and Intelligent Technology, Ministry of Education, Guangxi Normal University(教育部教育区块链与智能技术重点实验室,广西师范大学);Department of Computer Science, University of Illinois at Chicago(伊利诺伊大学芝加哥分校计算机科学系)
CHoE: Cross-Domain Heterogeneous Graph Prompt Learning via Structure-Conditioned Experts
School of Computer Science and Technology, Tianjin University(天津大学计算机科学与技术学院);Department of Health Technology and Informatics, and Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University(香港理工大学健康科技与信息学系、数据科学与人工智能系)
Controllable Image Captioning with Prompt-Conditioned Scene Rewards
POSTECH(浦项科技大学);Graduate School of Artificial Intelligence, POSTECH(浦项科技大学人工智能研究生院);Department of Computer Science and Engineering, POSTECH(浦项科技大学计算机科学与工程系)
Turn-Based Structural Triggers: Structure-Conditioned Backdoors in Multi-Turn LLMs
School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院);Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);Ant Group(蚂蚁集团)
Form, Not Content? A Preregistered, Placebo-Controlled Evaluation of Learned Error-Conditioned Self-Repair Through Prompts and Weights in Frozen Small Code Models
PythaLab, Yıldız Technical University, Istanbul, Turkey(PythaLab,Yıldız技术大学,伊斯坦布尔,土耳其)