The throughput gains of speculative decoding depend on the trade-off between acceptance length and execution latency. We ask when additional acceptance is worth the cost of obtaining it. We establish a comparison protocol spanning training, inference, and evaluation, remeasuring released checkpoints and retraining six...
Yong-Chao Long, Chun-Rui Zheng, Yuxuan Zhou et al.· 0 citations
Vision-language models face a fundamental scaling bottleneck: the number of visual tokens grows with both temporal duration and spatial resolution, making long-video understanding expensive for the vision encoder and the language model. Existing methods often compress visual tokens after dense encoding, creating a mism...
Yulong Liu, Xiao-Tian Han, Jun-Yuan Shang et al.· 0 citations
Existing long-video agents acquire evidence through one uniform behavior, ignoring whether the required evidence is concentrated, requires broad occurrence coverage, or must discriminate competing hypotheses---which can cause failure before substantive reasoning begins. Prescribing a fine-grained solution procedure for...
Can-Can Zhang, Bao-Feng Zhang, Xiao-Tian Han et al.· 0 citations
Aggregating Visual Information with Optimal Transport (AVIOT), which casts video token compression as transporting a dense empirical measure of frame observations onto a compact target measure, and adapts this construction along task and spatial axes to preserve visual information dispersed across frames under compress...
Wen-Ti Yin, Xiao-Tian Han, Jun-Yuan Shang et al.· 0 citations
This work proposes Autonomy-of-Heads (AoH), a data-free method that identifies retrieval and streaming heads from the spectral geometry of query-key projections that retains 96.5% of Full Attention performance on average while reducing prefill and decode latency by up to 41.4 and 66.0%, respectively, and KV-cache memor...
Ye-Han Yang, Jun-Yuan Shang, Yang Li et al.· 0 citations
This work presents a principled analysis of the distributions induced by lossy verification methods, and shows that many seemingly distinct approaches differ only superficially and can be unified into two categories: truncation-based verification and collaborative verification.
Tian-Yu Wang, Yuxuan Zhou, Heng Li et al.· arXiv.org· 0 citations
Memory-Augmented Compression is proposed, a training-free framework that constructs reusable reasoning memories from historical traces and retrieves them as prefill-side scaffolds to compensate for information lost during compression.
Si-Meng Zhang, Yi-Long Chen, Wenyuan Zhang et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
