arXiv:2508.03351v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated remarkable capabilities across diverse language tasks, motivating their extension to vision-la...
The paper introduces Orthogonal Drift Correction (ODC), an inference-time guidance method that mitigates semantic drift in text-to-image generation without retraining or extra user input. ODC operates in two stages: first it detects drift by comparing the generated image to the prompt in a shared vision-language embedding space, isolates the orthogonal component of the drift, and converts it back into text; second it refines the text conditioning for a second generation pass using an adaptive rank-reduced concept removal module. Experiments show ODC improves prompt-image alignment, raising CLIP scores by up to 5.4% on COALBench-300 and PartiPrompts, and boosting the 3-in-1 score by 4.5% on T2I-CompBench’s complex subset.
The paper surveys training‑free methods to accelerate inference in Video Large Language Models (VLMs), aiming to reduce computational, latency, and memory costs while retaining task‑relevant spatiotemporal evidence. It categorizes techniques by their intervention points in the inference pipeline—input, visual encoding, projector, prefill, and decoding—and discusses post‑training quantization as a cross‑stage approach. The authors analyze how the position of each intervention influences evidence preservation and computational savings, noting that local efficiency gains may not lead to overall acceleration, and they provide quantitative comparisons from prior work and new experiments to evaluate accuracy‑latency trade‑offs.
CoFiE introduces a two‑stage evidence selection framework for streaming video understanding, separating a coarse, query‑agnostic filtering of visually distinctive frames from a fine, query‑specific refinement during LLM prefill. By filtering out redundant frames before expensive vision encoding, CoFiE reduces end‑to‑end latency while maintaining high accuracy. The method achieves state‑of‑the‑art performance on benchmarks such as StreamingBench and OvO‑Bench, improving accuracy by up to 3.15% and inference speed by up to 2.54× compared to prior approaches.
arXiv:2609.16656v1 Announce Type: new Abstract: State space models (SSMs), particularly Mamba, have emerged as efficient alternatives to attention-based architectures and have been extended to vision...
arXiv:2609.26425v1 Announce Type: new Abstract: KV cache memory has become a major deployment bottleneck for video generation and world models, which motivates low-bit quantization study for efficien...
arXiv:2609.37042v1 Announce Type: cross Abstract: Video Large Language Models (VideoLLMs) have achieved strong video understanding capabilities but incur substantial inference overhead due to the lar...
arXiv:2608. 03112v1 Announce Type: cross Abstract: Vision-language models excel at image and video understanding but suffer from high inference latency due to the need to process thousands of tokens per image, limiting their deployment on resource-constrained edge devices and in real-time surveillance applications.
arXiv:2607.28627v2 Announce Type: replace-cross Abstract: Long visual contexts challenge vision-language models: performance degrades as the number of distractors grows, and processing all tokens at...
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.
LensVLM is an inference framework and post‑training recipe that lets Vision‑Language Models (VLMs) process compressed images of text by selectively expanding only the relevant parts back to full resolution. Using Qwen3.5‑9B‑Base, LensVLM achieves accuracy comparable to full‑text models at 4.3× compression and outperforms other compression baselines up to 10.1× across seven text QA benchmarks, while also improving performance on multimodal document and code tasks as compression increases.
arXiv:2512. 08240v2 Announce Type: replace-cross Abstract: Vision-language models (VLMs) rely on hundreds of visual tokens, leading to high computational and memory costs.