Few-shot in-context learning (ICL) with multi-modal large language models (MLLMs) enables task adaptation without parameter updates, but its performance is highly sensitive to the quality and coverage of the selected demonstrations. While unlabeled multi-modal data is abundant, it remains elusive how to exploit them fo...
Zirui Cheng, Xun Xu, Tiankai Chen et al.· 0 citations
TARL is introduced, a memory state update framework that maps each statement to one of five executable actions and is trained by comparing the memory states produced by alternative update operations, encouraging the model to select the operation that leads to the correct result.
Han Xiao, Hong-Jun Xu, Xin Zhang et al.· 1 citation
Experiments on both in-domain (MuST-C) and out-of-domain (Europarl-ST) En-De and En-Es datasets demonstrate that EASiST offers superior latency-quality trade-offs compared to several strong baselines.
Biao Fu, Donglei Yu, Min-Peng Liao et al.· AAAI Conference on Artificia...· 16 citations· ⚡1
A novel framework based on conditional Variational autoencoder for SLT (VSLT) that facilitates direct and sufficient cross-modal alignment between sign language videos and spoken language text is proposed, and a shared Attention Residual Gaussian Distribution (ARGD) which considers the textual information as a residual...
Rui Zhao, Liang Zhang, Biao Fu et al.· International Journal of Com...· 0 citations
Credit-Aware Hierarchical Memory Evolution (CHIME), a self-evolving memory framework that maintains a separate planning bank and execution bank and follows an attribute-before-memorize principle, which shows that CHIME consistently outperforms state-of-the-art training-based and self-evolving memory baselines.
Yongshi Ye, Tian Lan, Feihu Jiang et al.· 0 citations
This work proposes OM-GRPO, a label-free RLVR framework that decouples reward estimation from policy optimization and introduces Contrast-Augmented Reward, which refines reward estimation via low-cost pairwise comparisons over existing trajectories without additional rollouts.
Yongshi Ye, Liang Zhang, Yidong Chen et al.· 0 citations
TARL is introduced, a memory state update framework that maps each statement to one of five executable actions and is trained by comparing the memory states produced by alternative update operations, encouraging the model to select the operation that leads to the correct result.
Han Xiao, Hongjun Xu, Xin Zhang et al.· 0 citations
PAMT is proposed, a process-aligned training framework that combines cold-start domain-aware Long-CoT supervision with reinforcement learning that improves over base models, outperforms MT-specialized baselines on average, and remains competitive with strong LLMs/LRMs across in-domain, OOD, and multilingual settings.
Yongshi Ye, Biao Fu, Chongxuan Huang et al.· 0 citations
TwT (Translation with Thought), a resource-rational framework that learns to modulate inference between intuitive and deliberate reasoning, confirms that aligning translation behavior with cognitive principles enables robust generalization, high translation quality, and efficient reasoning in MDMT.
Yongshi Ye, Biao Fu, Chongxuan Huang et al.· Annual Meeting of the Associ...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
