Pith. sign in

REVIEW 4 cited by

Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.00557 v1 pith:ZWGVMJJ2 submitted 2025-04-01 cs.CV cs.LG

Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features

Jewon Lee , Ki-Ung Song , Seungmin Yang , Donguk Lim , Jaeyeon Kim , Wooksu Shin , Bo-Kyeong Kim , Yong Jae Lee
show 1 more author
Tae-Ho Kim
This is my paper
classification cs.CV cs.LG
keywords featuresvisualtokenscachecross-attentionimageinferencelayers
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Visual token reduction lowers inference costs caused by extensive image features in large vision-language models (LVLMs). Unlike relevant studies that prune tokens in self-attention-only LVLMs, our work uniquely addresses cross-attention-based models, which achieve superior performance. We identify that the key-value (KV) cache size for image tokens in cross-attention layers significantly exceeds that of text tokens in self-attention layers, posing a major compute bottleneck. To mitigate this issue, we exploit the sparse nature in cross-attention maps to selectively prune redundant visual features. Our Trimmed Llama effectively reduces KV cache demands without requiring additional training. By benefiting from 50%-reduced visual features, our model can reduce inference latency and memory usage while achieving benchmark parity.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety

    cs.CR 2026-04 unverdicted novelty 7.0

    ProjLens shows that backdoor parameters in MLLMs are encoded in low-rank subspaces of the projector and that embeddings shift toward the target direction with magnitude linear in input norm, activating only on poisone...

  2. Reward Valuation in Vision Language Models: Causal Mechanisms Underlying Anhedonia

    cs.LG 2026-07 conditional novelty 6.5

    Targeted perturbation of reward-anticipatory units in VLMs induces anhedonia-like effort avoidance and clinical-scale score drops without impairing baseline task competence.

  3. HotComment: A Benchmark for Evaluating Popularity of Online Comments

    cs.AI 2026-04 unverdicted novelty 6.0

    HotComment is a new multimodal benchmark that quantifies online comment popularity via content quality assessment, interaction-based prediction, and agent-simulated user engagement, accompanied by the StyleCmt stylist...

  4. C-CoT: Counterfactual Chain-of-Thought with Vision-Language Models for Safe Autonomous Driving

    cs.CV 2026-05 unverdicted novelty 5.0

    C-CoT applies VLMs to autonomous driving via five-stage reasoning with a meta-action tree for counterfactuals, yielding 81.9% risk recall, 3.52% collision rate, and 1.98 m L2 error on a new dataset.