AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

EgoAdapt: Adaptive Multisensory Distillation and Policy Learning for Efficient Egocentric Perception

Sanjoy Chowdhury, Subrata Biswas, Sayan Nag, Tushar Nagarajan, Calvin Murdock, Ishwarya Ananthabhotla, Yijun Qian, Vamsi Krishna Ithapu, Dinesh Manocha, Ruohan Gao

ICCV 2025 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 3 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 policy module TeMPLe on top of distillation (ablation)EgoAdapt · EPIC-Kitchensaction accuracy: 50.88 -> 56.74Improved
add policy module TeMPLe on top of distillation (ablation)EgoAdapt · EasyCom ASLmAP: 78.59 -> 89.74Improved
use learned policy instead of random policy (ablation)CFD · EasyCom ASLmAP: 67.41 -> 89.74Improved
use learned policy instead of heuristic selection (ablation)CFD · EasyCom ASLmAP: 65.88 -> 89.74Improved
add stage 3 joint finetuning (ablation)EgoAdapt · EasyCom ASLmAP: 83.64 -> 89.74Improved
policy-driven modality selection instead of all modalities (ablation)EgoAdapt · EasyCom ASLmAP and GMACs: 88.07 -> 89.74 mAP, 0.148 -> 0.073 GMACsImproved
add visual modality to 4-channel audio (ablation)fixed fusion · EasyCom ASLmAP: 81.27 -> 83.44Improved
learned frame selection instead of random frame (ablation)EgoAdapt · EPIC-Kitchensaction accuracy: 48.63 -> 56.74Improved
learned frame selection instead of using all frames (ablation)EgoAdapt · EPIC-KitchensGMACs: 16.51 -> 7.14 GMACsImproved
increase quantization from 4-bit to 32-bitEgoAdapt · EasyCom ASLmAP: 78.92 -> 89.74Improved
swap teacher from TIM to MBTEgoAdapt · EPIC-Kitchensaction accuracy: 56.74 -> 42.53Hurt
swap teacher from MUST to GLCEgoAdapt · AEA behavior anticipationgaze MAE T700ms: 14.58 -> 19.32Hurt

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