You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Access is granted manually by the owner. State your name, affiliation and intended use.

VideoFace-X-LLM — ViT-only fine-tune (step 3406 of 3406)

InternVideo2.5-Chat-8B fine-tuned on the ExplainFace corpus (Gemini descriptions of CelebV-HQ videos) for the ACCV 2026 paper "When, Where, and Why: Hierarchical Causal Explainability for Video Face Understanding with MLLMs". This is the "Ours (ViT-Only)" row of the results table.

Training

Trainable: vision encoder (lr 1e-5) + connector mlp1 (lr 4e-5); language model frozen.
1 epoch = 3,406 steps over the 27,304-clip identity-independent training split (Aak579/CelebV-HQ-identity-split20), micro-batch 1, global batch 8, AdamW, weight decay 0.01, 3 % linear warm-up then cosine decay, gradient clip 1.0, 8–32 frames per clip, seed 42, FSDP full sharding (xtuner unify_internvl2_train_r16.py). Full args: train_run_config.json.
This checkpoint is the end of the epoch.

Test results (6,923-clip identity-independent test split)

Appearance mA / F1mac	Action mA / F1mac	Emotion video Acc / F1mac	Emotion temporal Acc / F1mac	Deepfake (FF++) Acc / F1mac	AU (DISFA) mA / frame F1 / clip F1
86.16 / 26.29	94.50 / 17.88	55.61 / 22.00	71.47 / 23.83	50.00 / 35.23	71.0 / 16.7 / 16.8

Loading

The folder is a full HF checkpoint (trust_remote_code=True; the modeling_*.py / conversation.py files are copied from the base snapshot). For the JSON tasks the model was decoded with the training-style prompt prefix and a { prefill, as in the paper's inference scripts.

Downloads last month: -

Safetensors

Model size

8B params

Tensor type

BF16

Inference Providers NEW

Video-Text-to-Text

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Aak579/VideoFace-X-LLM-ViT-only-3406

Base model

OpenGVLab/InternVideo2_5_Chat_8B

Finetuned

(3)

this model