Results
Offline full-set eval (VideoMME-v1 2700 + PerceptionComp 1108 + Video-Holmes 1837 = 5645 rows), scored with the repo's vero compute_score. mean = macro-mean of the 3 bench accuracies.
Keeper = global_step_80 (peak):
Table with columns: metric, mean, videomme, holmes, perceptioncomp| metric | mean | videomme | holmes | perceptioncomp |
|---|
| τ0.8 n16/k8 @80 (keeper) | 0.4913 | 0.653 | 0.471 | 0.349 |
| cold-base RL keeper @80 | 0.4918 | 0.658 | 0.474 | 0.343 |
| τ0.95 sibling peak (n8/k4) | 0.4881 | — | — | — |
| τ0.8 sibling peak (n8/k4) | 0.4864 | — | — | — |
| stock Qwen3-VL-8B ckpt-0 (zero-shot) | 0.4444 | 0.6426 | 0.4143 | 0.2762 |
Trajectory (mean): 0.488 @20 → 0.490 @40 → 0.487 @60 → 0.491 @80 → clean monotone fade (0.481 @120, 0.473 @140), no collapse. (Step 100 eval skipped — reproducibly hung on one deterministic shard, step-100-specific not systemic.) Peak edges the n8 explore siblings by ~0.3 pt = within noise.
Training
- Base model:
Qwen/Qwen3-VL-8B-Instruct, warm-started from the SFT-770-RL run's global_step_60 (merged HF _merged/sft770_perf_gs60_hf, itself val 0.4814). Lineage: stock 8B → OMR-SFT-770 → video-RL gs60 → this run.
- Framework: fork of volcengine/verl —
ngquangtrung57/verl@videorl-mods. Fully-async GRPO: FSDP2 trainer + vLLM rollouter.
- Reward: dapo-style
score = 0.8·accuracy + 0.2·format (FORMAT_WEIGHT=0.2, FORMAT_MIN_THINK_CHARS=100). No KL penalty.
- Topology: 4-node 2+2 — 2 trainer nodes (16-GPU FSDP2, dp=16) + 2 rollout nodes (16 GPU, vLLM TP=2 → 8 replicas). H100×8 per node.
- × × = (vs 512 at n=8; bigger GRPO groups = 8 explore + 8 anchor per selected prompt).
Table with columns: key, value| key | value |
|---|
enable | true |
trigger_mode | high |
top_prob_threshold (τ) | 0.8 |
k_explore | 8 (of n=16; 8 explore + 8 anchor) |
prompt_exploration_prob | 0.5 |
deterministic |
- Validation: inline val OFF (
test_freq=10000); video val is offline full-set eval on a dedicated H100 node (vLLM TP1), every 20 fit-steps.
- Train metrics: ~445 s/step (vs ~290 at n=8 — the bigger groups cost ~55% more wall-clock); reward → ~0.88; response_length ~250 tok; format ~1.0; no OOM/collapse. Stopped by user at fit-step ~106/781 (verdict clear).
W&B
Project verl_fully_async (entity quangtrung5705-nanyang-technological-university-singapore). The logical run has three crash-resume segments (vtlkaq8u, iwwcf7do, hb4vi5h3); the last/most-complete (hb4vi5h3) carries the curated label. Train metrics only — video val is offline, not on W&B:
https://wandb.ai/quangtrung5705-nanyang-technological-university-singapore/verl_fully_async/runs/hb4vi5h3
Intended use / limitations
Research checkpoint — the group-size ablation arm of an 8B video study. Headline finding: bigger GRPO groups (n16/k8) do not break the ~0.485 video dead-heat (peak 0.4913 ties the best keeper then fades). With this run, seven levers (SFT warmstart, cold-base RL, cold exploration, base-warm explore, SFT-warm explore τ0.95, SFT-warm explore τ0.8, and this τ0.8 n16/k8) all peak ~0.485-0.49 — the video ceiling is bound by something outside the exploration / warm-start / RL-duration / τ / group-size space. Multiple-choice video QA, <think>…</think> then-answer format. No safety/RLHF alignment beyond the base.
Usage
from transformers import AutoModelForImageTextToText, AutoProcessor
model_id = "ngqtrung/video-8b-grpo-ppexplore-n16k8"
model = AutoModelForImageTextToText.from_pretrained(model_id, dtype="auto", device_map="auto")
processor = AutoProcessor.from_pretrained(model_id)
messages = [{
"role": "user",
"content": [
{"type": "video", "video": "clip.mp4"},
{"type": "text", "text": "Answer the multiple-choice question. Reason inside <think>...</think>, then give the final letter."},
],
}]
inputs = processor.apply_chat_template(
messages, add_generation_prompt=True, tokenize=True,
return_dict=True, return_tensors="pt"
).to(model.device)
out = model.generate(**inputs, max_new_tokens=1024)
print(processor.batch_decode(out[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)[0])
Citation / lineage
- Base model: Qwen3-VL-8B-Instruct (Qwen team). Inherits the Qwen3-VL license — review the base model's terms; the Apache-2.0 tag refers to this repo's RLVR training artifacts.
- Warm start: SFT-770-RL
global_step_60 (lineage: stock 8B → OMR-SFT-770 → video-RL gs60).
- Framework: verl (volcengine/verl), fork
ngquangtrung57/verl@videorl-mods; fully-async GRPO (FSDP2 + vLLM).
- Method: entropy-aware token-dropout exploration ("ppexplore", τ=0.8) with enlarged GRPO groups (n=16, k_explore=8). Group-size ablation in a controlled OMR/Video exploration study on Qwen3-VL-8B (
docs/experiments_summary_8b.md).