7 papers
FrameOracle: Learning What to See and How Much to See in Videos
Chaoyu Li, Tianzhi Li, Fei Tao +6
Vision-language models (VLMs) advance video understanding but operate under tight computational budgets, making performance dependent on selecting a small, high-quality subset of f…
OpenGenAlign: A Preference Dataset and Benchmark for Trustworthy Reward Modeling in Open-Ended, Long-Context Generation
Hanning Zhang, Juntong Song, Juno Zhu +3
Reward Modeling is critical in evaluating and improving the generation of Large Language Models (LLMs). While numerous recent works have shown its feasibility in improving safety,…
DMOSpeech 2: Reinforcement Learning for Duration Prediction in Metric-Optimized Speech Synthesis
Yinghao Aaron Li, Xilin Jiang, Fei Tao +4
Diffusion-based text-to-speech (TTS) systems have made remarkable progress in zero-shot speech synthesis, yet optimizing all components for perceptual metrics remains challenging.…
DuaShepherd: Integrating Stepwise Correctness and Potential Rewards for Mathematical Reasoning
Yuanhao Wu, Juntong Song, Hanning Zhang +2
In this paper, we propose DuaShepherd, a novel reward modeling framework that integrates two complementary reward signals, correctness and potential, to enhance the mathematical re…
VeraCT Scan: Retrieval-Augmented Fake News Detection with Justifiable Reasoning
Cheng Niu, Yang Guan, Yuanhao Wu +7
The proliferation of fake news poses a significant threat not only by disseminating misleading information but also by undermining the very foundations of democracy. The recent adv…
Enhancing Dialogue State Tracking Models through LLM-backed User-Agents Simulation
Cheng Niu, Xingguang Wang, Xuxin Cheng +2
Dialogue State Tracking (DST) is designed to monitor the evolving dialogue state in the conversations and plays a pivotal role in developing task-oriented dialogue systems. However…