AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

STEP: Spatial Thinking and Egocentric Pointing for Embodied Instruction Following

Hanxuan Li, Bin Fu, Zeyuan Lin, Ruiping Wang, Xilin CHEN

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add RGB action data fine-tuning (ablation)Qwen2.5-VL-7B · ALFRED Val UnseenSR: 7.87 -> 13.92Improved
add Set-of-Mark traversable-region markers (ablation)Qwen2.5-VL-7B · ALFRED Val UnseenSR: 13.92 -> 18.43Improved
add BEV map modality via simple concatenation (ablation)Qwen2.5-VL-7B · ALFRED Val UnseenSR: 18.43 -> 19.22Improved
add map modality (ablation)Qwen2.5-VL-7B · ALFRED Val UnseenGC: 30.61 -> 29.85No clear effect
add STEP-CoT reasoning traces (ablation)STEP · ALFRED Val UnseenSR: 19.22 -> 23.72Improved
add STEP-CoT reasoning traces (ablation)STEP · ALFRED Val SeenSR: 21.12 -> 28.40Improved
add primitive-skill data (ablation)STEP · ALFRED Val UnseenGC: 33.33 -> 39.36Improved
add primitive-skill data (ablation)STEP · ALFRED Val SeenGC: 40.35 -> 39.69Hurt
add primitive-skill data (ablation)STEP · ALFRED Val UnseenSR: 23.72 -> 26.77Improved
provide oracle segmentation (ablation)STEP-7B · ALFRED Val UnseenSR: 26.77 -> 41.57Improved
provide oracle goal decomposition on top of oracle segmentation (ablation)STEP-7B · ALFRED Val UnseenSR: 41.57 -> 55.69Improved
fine-tune on STEP-CoT vs zero-shot GPT-4oSTEP-7B · ALFRED Val UnseenSR: 16.00 -> 26.77Improved

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