AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

STEP-Nav: Spatial-Temporal Efficient Visual Token Pruning for Vision-and-Language Navigation with Large Language Models

Yantao Lu, Shiqi Sun, Ning Liu, Bo Jiang, Ying Zhang, Jinchao Chen, Chenglie Du

AAAI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes STEP-Nav, a spatial-temporal visual token pruning method to reduce the computational cost of LLM-based vision-and-language navigation.

Problem

LLM-based vision-and-language navigation encodes image sequences as dense visual tokens, incurring substantial computational overhead.

Approach

Prunes visual tokens across both spatial and temporal dimensions of the navigation image sequence to cut LLM input cost while preserving navigation accuracy (based on the title; details not available in the abstract provided).

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use STEP-Nav instead of ToMe at 88.9% pruningNavid-CE · VLN-CE R2R val-unseenOS: 36.8 -> 46.2Improved
use STEP-Nav instead of ToMe at 88.9% pruningNavid-CE · VLN-CE R2R val-unseenSPL: 25.9 -> 33.2Improved
use STEP-Nav instead of VisionZip at 88.9% pruningNavid-CE · VLN-CE R2R val-unseenSPL and SR: +0.3% SPL, +2.5% SRImproved
prune 66.7% of tokens with STEP-Nav vs full tokensNavid-CE · VLN-CE R2R val-unseenSR: 37.4 -> 36.4No clear effect
prune 88.9% of tokens with STEP-Nav vs full tokensNavid-CE · VLN-CE R2R val-unseenSPL: 35.9 -> 33.2Hurt
use STEP-Nav instead of ToMe at 88.9% pruningNavGPT-2 FlanT5-XL · R2R val-unseenSR: 52.6 -> 65.6Improved
use STEP-Nav instead of VisionZip at 88.9% pruningNavGPT-2 FlanT5-XL · R2R val-unseenSR: 64.2 -> 65.6Improved
use STEP-Nav instead of VisionZip at 88.9% pruningNavid-CE · VLN-CE RxR val-unseen (zero-shot)SR: 21.8 -> 22.4No clear effect
use STEP-Nav instead of ToMe at 88.9% pruningNavid-CE · VLN-CE RxR val-unseen (zero-shot)SR: 17.4 -> 22.4Improved
use STEP-Nav instead of VisionZip at 85% pruningNavGPT-2 FlanT5-XL (1.5B) · R2R val-unseenSR: 64.5 -> 66.3Improved
use STEP-Nav instead of VisionZip at 85% pruningNavGPT-2 FlanT5-XXL (5B) · R2R val-unseenSR: 70.1 -> 69.6Hurt
use STEP-Nav instead of VisionZip at 77.8% pruningNavid-CE · Real-world (3 indoor environments)SR: 78% -> 80% meeting room; 67% -> 69% bedroom; 84% -> 91% corridorImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy