List your company

Goal reasoning on EgoToM

89AccuracyHumans
6.94428.24749.5570.853Mar 25, 2026

Evaluation results

MethodLinks
HumansSetting=Baseline, Nframe=last 5s2026.0389
Qwen2.5-VL-7BSetting=w/o \delta _{T,l}^h, Nframe=242026.0388.9
Qwen2.5-VL-7BSetting=+\alpha \Delta, Nframe=242026.0388.9
Qwen2.5-VL-7BSetting=w/o \delta _{V,l}^h, Nframe=242026.0388.4
HumansSetting=Baseline, Nframe=last 30s2026.0388
Qwen2.5-VL-7BSetting=Rnd-\Delta, Nframe=242026.0387.1
Qwen2.5-VL-7BSetting=Baseline, Nframe=242026.0386.9
Gemini-2.5-FlashSetting=Baseline, Nframe=242026.0386
Video-Llama2-72BSetting=Baseline, Nframe=82026.0385
GPT-4-TurboSetting=Baseline, Nframe=202026.0383
HumansSetting=Baseline, Nframe=last 30s2026.0378
HumansSetting=Baseline, Nframe=last 5s2026.0377
LLaVA-Next-Video-7BSetting=+\alpha \Delta, Nframe=242026.0374.5
LLaVA-Next-Video-7BSetting=w/o \delta _{T,l}^h, Nframe=242026.0373.2
CogVLM2Setting=Baseline, Nframe=242026.0373
LLaVA-Next-Video-7BSetting=w/o \delta _{V,l}^h, Nframe=242026.0372.6
GPT-4oSetting=Baseline, Nframe=242026.0368.7
LLaVA-Next-Video-7BSetting=Rnd-\Delta, Nframe=242026.0362.1
LLaVA-Next-Video-7BSetting=Baseline, Nframe=242026.0361.5
Qwen2.5-VL-7BSetting=-\alpha \Delta, Nframe=242026.0360.3
LLaVA-Next-Video-7BSetting=-\alpha \Delta, Nframe=242026.0350.6
GPT-4-TurboSetting=Baseline, Nframe=202026.0342
Gemini-2.5-FlashSetting=Baseline, Nframe=242026.0340.1
Video-Llama2-72BSetting=Baseline, Nframe=82026.0340
Qwen2.5-VL-7BSetting=+\alpha \Delta, Nframe=242026.0337.6
Qwen2.5-VL-7BSetting=w/o \delta _{V,l}^h, Nframe=242026.0337.5
CogVLM2Setting=Baseline, Nframe=242026.0336
Qwen2.5-VL-7BSetting=w/o \delta _{T,l}^h, Nframe=242026.0333.2
Qwen2.5-VL-7BSetting=Rnd-\Delta, Nframe=242026.0331.9
Qwen2.5-VL-7BSetting=Baseline, Nframe=242026.0331.1
LLaVA-Next-Video-7BSetting=+\alpha \Delta, Nframe=242026.0329.7
LLaVA-Next-Video-7BSetting=w/o \delta _{V,l}^h, Nframe=242026.0329
LLaVA-Next-Video-7BSetting=Rnd-\Delta, Nframe=242026.0325.4
LLaVA-Next-Video-7BSetting=w/o \delta _{T,l}^h, Nframe=242026.0325.3
LLaVA-Next-Video-7BSetting=Baseline, Nframe=242026.0324
GPT-4oSetting=Baseline, Nframe=242026.0322.6
Qwen2.5-VL-7BSetting=-\alpha \Delta, Nframe=242026.0314.3
LLaVA-Next-Video-7BSetting=-\alpha \Delta, Nframe=242026.0310.1