English
Related papers

Related papers: Unhackable Temporal Rewarding for Scalable Video M…

200 papers

We provide the first formal definition of reward hacking, a phenomenon where optimizing an imperfect proxy reward function leads to poor performance according to the true reward function. We say that a proxy is unhackable if increasing the…

Machine Learning · Computer Science 2025-03-07 Joar Skalse , Nikolaus H. R. Howe , Dmitrii Krasheninnikov , David Krueger

Reinforcement learning (RL) systems typically optimize scalar reward functions that assume precise and reliable evaluation of outcomes. However, real-world objectives--especially those derived from human preferences--are often uncertain,…

Machine Learning · Computer Science 2026-04-30 Disha Singha

Dense video captioning aims to interpret and describe all temporally localized events throughout an input video. Recent state-of-the-art methods leverage large language models (LLMs) to provide detailed moment descriptions for video data.…

Computer Vision and Pattern Recognition · Computer Science 2026-01-07 Wei-Yuan Cheng , Kai-Po Chang , Chi-Pin Huang , Fu-En Yang , Yu-Chiang Frank Wang

Compared with static knowledge graphs, temporal knowledge graphs (tKG), which can capture the evolution and change of information over time, are more realistic and general. However, due to the complexity that the notion of time introduces…

Computation and Language · Computer Science 2025-04-07 Siheng Xiong , Yuan Yang , Faramarz Fekri , James Clayton Kerce

The successes of reinforcement learning in recent years are underpinned by the characterization of suitable reward functions. However, in settings where such rewards are non-intuitive, difficult to define, or otherwise error-prone in their…

Formal Languages and Automata Theory · Computer Science 2023-03-02 Mohammad Afzal , Sankalp Gambhir , Ashutosh Gupta , Krishna S , Ashutosh Trivedi , Alvaro Velasquez

Multimodal Large Language Models (MLLMs) have shown strong performance in video understanding tasks. However, they continue to struggle with long-form videos because of an inefficient perception of temporal intervals. Unlike humans, who can…

Computer Vision and Pattern Recognition · Computer Science 2025-07-08 Chenglin Li , Qianglong Chen , fengtao , Yin Zhang

The pervasive deployment of large language models (LLMs) in conversational AI systems has revolutionized information access, yet their propensity for generating factually unsupported or hallucinated responses remains a critical impediment…

Computation and Language · Computer Science 2025-06-03 Steven Robinson , Antonio Carlos Rivera

The Arrow of Time (AoT)-time's irreversible flow shaping physical events-is fundamental to video comprehension, yet remains a significant challenge for modern large multimodal models (LMMs). Current LMMs struggle to perceive and utilize…

Computer Vision and Pattern Recognition · Computer Science 2025-10-27 Zihui Xue , Mi Luo , Kristen Grauman

Reward models in RLHF are trained to score only the final token of a response - a choice that discards rich signal from every intermediate position and produces models whose token-level outputs are noise. We argue this is a missed…

Machine Learning · Computer Science 2026-04-28 Alex Nikulkov

Reward hacking arises when a model improves a proxy reward by exploiting shortcuts rather than solving the intended task. We study this failure mode through the geometry of reinforcement learning updates in language models and argue that…

Machine Learning · Computer Science 2026-05-26 Wenlong Deng , Jiaji Huang , Kaan Ozkara , Yushu Li , Christos Thrampoulidis , Xiaoxiao Li , Youngsuk Park

Aligning generative diffusion models with human preferences via reinforcement learning (RL) is critical yet challenging. Most existing algorithms are often vulnerable to reward hacking, such as quality degradation, over-stylization, or…

Reward models are central to both reinforcement learning (RL) with language models and inference-time verification. However, existing reward models often lack temporal consistency, leading to ineffective policy updates and unstable RL…

Machine Learning · Computer Science 2025-09-30 Dan Zhang , Min Cai , Jonathan Light , Ziniu Hu , Yisong Yue , Jie Tang

This paper explores continuous-time control synthesis for target-driven navigation to satisfy complex high-level tasks expressed as linear temporal logic (LTL). We propose a model-free framework using deep reinforcement learning (DRL) where…

Robotics · Computer Science 2023-03-17 Mingyu Cai , Makai Mann , Zachary Serlin , Kevin Leahy , Cristian-Ioan Vasile

Recent years have seen considerable research activities devoted to video enhancement that simultaneously increases temporal frame rate and spatial resolution. However, the existing methods either fail to explore the intrinsic relationship…

Computer Vision and Pattern Recognition · Computer Science 2021-09-02 Zhihao Shi , Xiaohong Liu , Chengqi Li , Linhui Dai , Jun Chen , Timothy N. Davidson , Jiying Zhao

Linear temporal logic (LTL) is a specification language for finite sequences (called traces) widely used in program verification, motion planning in robotics, process mining, and many other areas. We consider the problem of learning LTL…

Artificial Intelligence · Computer Science 2026-01-22 Ritam Raha , Rajarshi Roy , Nathanaël Fijalkow , Daniel Neider

This paper addresses the problem of learning optimal control policies for systems with uncertain dynamics and high-level control objectives specified as Linear Temporal Logic (LTL) formulas. Uncertainty is considered in the workspace…

Robotics · Computer Science 2024-10-17 Yiannis Kantaros , Jun Wang

Deep Reinforcement Learning (DRL) has shown outstanding performance on inducing effective action policies that maximize expected long-term return on many complex tasks. Much of DRL work has been focused on sequences of events with discrete…

Machine Learning · Computer Science 2021-05-07 Yeo Jin Kim , Min Chi

Reinforcement Learning from Human Feedback (RLHF) has been crucial to the recent success of Large Language Models (LLMs), however, it is often a complex and brittle process. In the classical RLHF framework, a reward model is first trained…

Machine Learning · Computer Science 2024-11-06 Rafael Rafailov , Yaswanth Chittepu , Ryan Park , Harshit Sikchi , Joey Hejna , Bradley Knox , Chelsea Finn , Scott Niekum

Time series analysis provides essential insights for real-world system dynamics and informs downstream decision-making, yet most existing methods often overlook the rich contextual signals present in auxiliary modalities. To bridge this…

Machine Learning · Computer Science 2026-03-24 Yushan Jiang , Wenchao Yu , Geon Lee , Dongjin Song , Kijung Shin , Wei Cheng , Yanchi Liu , Haifeng Chen

Reinforcement learning (RL) has shown strong potential for enhancing reasoning in multimodal large language models, yet existing video reasoning methods often rely on coarse sequence-level rewards or single-factor token selection,…

Computer Vision and Pattern Recognition · Computer Science 2026-01-28 Ziyue Wang , Sheng Jin , Zhongrong Zuo , Jiawei Wu , Han Qiu , Qi She , Hao Zhang , Xudong Jiang