20 results for “cross-attention layers”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
Nityanand Mathur, Hamees Sayed, Wasim Madha, Apoorv Singh +3 more
This paper proposes a method for analyzing how individual words in style captions influence acoustic output in text-to-speech systems using cross-attention attribution for speech diffusion models.
Timing Yang, Predrag Neskovic, Jansen Seheult, Wenchao Han +3 more
This paper proposes a self-supervised visual model called RATS that discovers compositional structure in images.
This paper proposes a unified multi-task framework for handling six different entity pair relationship types in e-commerce with a shared relevance model.
Yutao Sun, Yanqi Zhang, Li Dong, Jianyong Wang +1 more
The paper proposes Cross-Layer Sparse Attention (CLSA) to significantly improve the efficiency and accuracy of long-context LLMs by jointly optimizing KV-cache sharing and the routing index across dec…
Yizhuo Lu, Changde Du, Qingyu Shi, Hang Chen +4 more
Mind-Omni introduces a unified multi-task framework that models the interplay between brain, vision, and language signals using a discrete diffusion paradigm, achieving state-of-the-art performance ac…
Kaixiong Gong, Xin Cai, Bin Lin, Hao Wang +8 more
This paper proposes Twins, a unified continuous token space for multimodal models using ViT and VAE features, and addresses optimization imbalance with a focal regression objective.
The paper analyzes token reduction for efficient unified VLM training, finding that while task-specific acceleration saves computation, it destroys the mutual performance gains achieved through joint…
This paper investigates the behavior of cross-encoders in information retrieval and explores their relevance prediction process, particularly in out-of-domain scenarios.
CORE-MTL proposes a representation-centric framework that uses causal orthogonal representations to disentangle task-relevant structure from nuisance variation in multi-task learning, achieving superi…
Shengkui Zhao, Zexu Pan, Haoxu Wang, Biao Tian +2 more
This paper proposes TF-MossFormer, a time-frequency transformer for monaural speech separation that combines local and global attention using a content-aware sliding-window mechanism.
This paper explores flexible routing in transformers for improved performance in LLMs.
Bohan Yang, Yijun Gong, Zhi Zhang, Ge Zhang +2 more
TriLens is a white-box detector that monitors the entropy of three internal streams (attention, feed-forward, residual) at every layer of a language model to detect hallucinations by tracking how inte…
The paper introduces Brain-IT-VQA, a novel framework that significantly improves visual question answering from fMRI signals, and presents NSD-VQA, a new, highly controlled dataset for this task.
The paper proposes UF-AMA, a unified framework that achieves state-of-the-art cross-domain emotion recognition by adaptively aligning and fusing multimodal physiological signals like EEG and eye-track…