arXiv AI
Jun 19
The Hidden Evolution of Disguised Visual Context inside the VLM
arXiv:2606. 20077v1 Announce Type: cross Abstract: Visual tokens enter Large Language Models (LLMs) as raw, foreign signals.
Dynamic Context Adapters: Efficiently Infusing History into Vision-and-Language Models
arXiv:2608. 10525v1 Announce Type: cross Abstract: Historical context integration presents a fundamental challenge for Vision-Language Models (VLMs) in sequential decision-making tasks.
Related stories
arXiv AI
Jul 14
Extending LLM Context via Associative Recurrent Memory
arXiv:2607. 11614v1 Announce Type: cross Abstract: Extending the context length of large language models (LLMs) is critical for many real-world applications, yet standard transformers remain constrained by quadratic compute and linear memory scaling.
arXiv Computer Vision
2d ago
FlashBack: Knowing When to Remember in Streaming Vision-Language Models
arXiv:2610.01192v1 Announce Type: new Abstract: Streaming vision-language models must process continuously growing video streams under a bounded compute budget, creating a persistent tension between...
arXiv Machine Learning
Sep 10
SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models
arXiv:2609.05533v1 Announce Type: cross Abstract: Long-horizon manipulation is partially observable: the information needed to choose the next action may appear only in observations from minutes earl...
arXiv Computer Vision
2d ago
Compressing History into Memory: Distilling Transformers into Recurrent Transformers
arXiv:2606.21562v2 Announce Type: replace Abstract: Transformers are AI's workhorse but their computational cost becomes prohibitive when processing long sequences. We target long-horizon streaming v...
arXiv AI
Aug 26
VisCache: Visual KV Cache Pruning for Efficient Vision Large Language Model Inference
VisCache introduces a two-stage, plug‑and‑play framework for pruning visual key‑value caches in Vision Large Language Models without retraining. The first stage filters out temporally redundant keyframes, while the second stage, PruneKV, applies a parabolic layer‑wise budget and asymmetric update to selectively prune keys and fuse values, preserving essential context. Experiments show up to 2.35× speedup and significant memory savings with only 19–28% of the original cache retained, outperforming existing baselines.
arXiv Computer Vision
Aug 25
UniMem: Unifying Multimodal Memory and Control for Vision-Language-Action Models
arXiv:2608.22869v1 Announce Type: cross Abstract: While Vision-Language-Action (VLA) models have leveraged internet-scale pretraining and task-focused finetuning to achieve strong performance on long...
arXiv Machine Learning
Jul 14
Controllably Efficient Language Models
arXiv:2511. 05313v2 Announce Type: replace Abstract: The substantial inference costs of attention in transformers motivated the development of efficient sequence mixers: namely sparse and sliding window attention, convolutions and linear attention.
arXiv Computer Vision
Sep 22
PREM: Prefix-Steered Recurrent Memory for Long-Video Understanding
arXiv:2609.23601v1 Announce Type: new Abstract: Long-video understanding must capture transient visual evidence under strict token budgets, yet existing methods compress frames, append memory tokens,...
arXiv AI
4d ago
Compress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation
The paper introduces Prediction‑Aligned Context Compaction (PACC), a method that learns a compact memory representation for long‑video generation by distilling a frozen video generator. PACC trains a compressor to aggregate past frames into memory tokens, using the generator as both teacher and student during on‑policy distillation. Experiments on MBench and VBench‑Long show that PACC improves memory‑event coverage and consistency, achieving better scores than strong baselines and producing competitive minute‑long videos.
arXiv AI
Aug 28
PACE: A Unified Condense-and-Extract Paradigm for Fast VLM Inference
PACE introduces a training‑free Condense‑and‑Extract framework that speeds up Vision‑Language Model inference by first adaptively downsampling visual inputs before encoding and then selectively retaining essential tokens during decoding. The Adaptive Pixel Compressor (APC) reduces encoder workload while preserving global context, and the Dynamic Dual‑Attention Extractor (DDAE) keeps task‑critical details by fusing visual and language signals. Applied to Qwen2.5‑VL‑7B, PACE maintains 93.8% of performance using only 10% of visual tokens, achieving a 3.1× speedup in time to first token.
arXiv AI
Jun 11
Task-Aware Structured Memory for Dynamic Multi-modal In-Context Learning
arXiv:2606. 11853v1 Announce Type: cross Abstract: Multi-modal large language models (MLLMs) depend on in-context learning (ICL) for rapid task adaptation, but their scalability is severely limited by finite context windows and the growing cost of key-value (KV) caches in long multi-modal sequences.