Human-focused
Collection
Our datasets that specifically have a real-world human as the subject in every image • 5 items • Updated • 1
Wang Weiyi PRO
kaupane
AI & ML interests
None yet
Recent Activity
upvoted a collection about 2 hours ago
Human-focused upvoted a paper 6 days ago
SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science? reacted to nwaughachukwuma's post with 👍 25 days ago
Can a text-only model + a vision toolkit (mm-ctx) match a native vision model?
We benchmarked 4 setups on 23 multimodal tasks (image, video, audio, PDF):
• glm-5.2 (text-only) + mm-ctx: 88.4
• gemini-3.5-flash (vision): 83
• deepseek-v4-pro (text-only) + mm-ctx: 79.4
• qwen3.6-35b-a3b (vision): 44.3
The best text-only setup `glm-5.2 + mm` outperformed gemini-3.5-flash, the top vision model, by 5.4 points (6.5%). It was also:
• 1.5x faster (100s vs 150s mean per task)
• the only setup with zero timeouts (46/46 completed; gemini timed out 4x on bulk-image and long-video tasks)
• the only setup stable across runs (88.5 / 88.4)
• top on video (100.0), image (91.7), and PDF (90.0) tasks
The trade-offs: the toolkit consumed 3.3x more tokens (4.25M vs 1.28M), and lost on audio (85.6 vs 71.3).
On completed tasks alone the two are nearly identical (91.0 vs 88.4): the toolkit's edge is efficient extraction that keeps long media tasks inside the time budget.
Full report: https://huggingface.co/blog/vlm-run/text-only-models-with-mmOrganizations
None yet
upvoted a collection about 2 hours ago
upvoted a paper 6 days ago
reacted to nwaughachukwuma's post with 👍 25 days ago
Post
675
Can a text-only model + a vision toolkit (mm-ctx) match a native vision model?
We benchmarked 4 setups on 23 multimodal tasks (image, video, audio, PDF):
• glm-5.2 (text-only) + mm-ctx: 88.4
• gemini-3.5-flash (vision): 83
• deepseek-v4-pro (text-only) + mm-ctx: 79.4
• qwen3.6-35b-a3b (vision): 44.3
The best text-only setup
• 1.5x faster (100s vs 150s mean per task)
• the only setup with zero timeouts (46/46 completed; gemini timed out 4x on bulk-image and long-video tasks)
• the only setup stable across runs (88.5 / 88.4)
• top on video (100.0), image (91.7), and PDF (90.0) tasks
The trade-offs: the toolkit consumed 3.3x more tokens (4.25M vs 1.28M), and lost on audio (85.6 vs 71.3).
On completed tasks alone the two are nearly identical (91.0 vs 88.4): the toolkit's edge is efficient extraction that keeps long media tasks inside the time budget.
Full report: https://huggingface.co/blog/vlm-run/text-only-models-with-mm
We benchmarked 4 setups on 23 multimodal tasks (image, video, audio, PDF):
• glm-5.2 (text-only) + mm-ctx: 88.4
• gemini-3.5-flash (vision): 83
• deepseek-v4-pro (text-only) + mm-ctx: 79.4
• qwen3.6-35b-a3b (vision): 44.3
The best text-only setup
glm-5.2 + mm outperformed gemini-3.5-flash, the top vision model, by 5.4 points (6.5%). It was also:• 1.5x faster (100s vs 150s mean per task)
• the only setup with zero timeouts (46/46 completed; gemini timed out 4x on bulk-image and long-video tasks)
• the only setup stable across runs (88.5 / 88.4)
• top on video (100.0), image (91.7), and PDF (90.0) tasks
The trade-offs: the toolkit consumed 3.3x more tokens (4.25M vs 1.28M), and lost on audio (85.6 vs 71.3).
On completed tasks alone the two are nearly identical (91.0 vs 88.4): the toolkit's edge is efficient extraction that keeps long media tasks inside the time budget.
Full report: https://huggingface.co/blog/vlm-run/text-only-models-with-mm
- 8 replies
liked a
model about 1 month ago
liked a
Space about 1 month ago
liked a
model 2 months ago
upvoted an article 3 months ago
Article
Towards Speed-of-Light Text Generation with Nemotron-Labs Diffusion Language Models
nvidia
• • 35 liked 4
models 4 months ago
XiaomiMiMo/MiMo-V2.5-Pro
Text Generation • 1T • Updated • 53.5k • • 743
openai/privacy-filter
Token Classification • 1B • Updated • 409k • • 1.73k
deepseek-ai/DeepSeek-V4-Pro
Text Generation • 1.6T • Updated • 980k • • 5.48k
OpenMOSS-Team/MOSS-Music-8B-Instruct
Audio-Text-to-Text • 9B • Updated • 1.48k • 31
updated a
dataset 4 months ago
published a
dataset 4 months ago
liked a
model 5 months ago
updated a
dataset 5 months ago
liked a
model 5 months ago
published a
dataset 5 months ago
liked a
dataset 5 months ago
reacted to SeaWolf-AI's post with 🔥 5 months ago
Post
4708
🌍 World Model Bench — does your world model actually think?
FID measures realism. FVD measures smoothness. But neither tells you whether the model understood the scene.
We just released WM Bench — the first benchmark for cognitive intelligence in world models. The core question: when a beast charges from 3 meters away, does the model know to sprint — not walk? Does it respond differently to a human vs an animal? Does it remember the left corridor was blocked two steps ago?
Those are cognitive questions. No existing benchmark asks them. So we built one.
3 Pillars · 10 Categories · 100 Scenarios · 1,000-point scale
- 👁 P1 Perception (25%) — Can it read the scene?
- 🧠 P2 Cognition (45%) — Does it predict threats, escalate emotions, utilize memory?
- 🔥 P3 Embodiment (30%) — Does the body respond with the right motion?
All evaluation is via simple JSON I/O — no 3D engine, no special hardware. Any model with an API can participate.
We also built PROMETHEUS as a live reference implementation — runs in your browser on a T4, no install needed. Combines FloodDiffusion motion generation with a LLM cognitive brain (Perceive → Predict → Decide → Act). Scored 726/1000 (Grade B) on Track C — the only directly verified model so far. Submissions from other teams very welcome.
---
🗂 Dataset → FINAL-Bench/World-Model
🌍 Demo → FINAL-Bench/World-Model
🏆 Leaderboard → FINAL-Bench/worldmodel-bench
📝 Article → https://huggingface.co/blog/FINAL-Bench/world-model
Part of the FINAL Bench Family — alongside FINAL Bench (Feb 2026). Feedback on rubrics and missing models always welcome!
FID measures realism. FVD measures smoothness. But neither tells you whether the model understood the scene.
We just released WM Bench — the first benchmark for cognitive intelligence in world models. The core question: when a beast charges from 3 meters away, does the model know to sprint — not walk? Does it respond differently to a human vs an animal? Does it remember the left corridor was blocked two steps ago?
Those are cognitive questions. No existing benchmark asks them. So we built one.
3 Pillars · 10 Categories · 100 Scenarios · 1,000-point scale
- 👁 P1 Perception (25%) — Can it read the scene?
- 🧠 P2 Cognition (45%) — Does it predict threats, escalate emotions, utilize memory?
- 🔥 P3 Embodiment (30%) — Does the body respond with the right motion?
All evaluation is via simple JSON I/O — no 3D engine, no special hardware. Any model with an API can participate.
We also built PROMETHEUS as a live reference implementation — runs in your browser on a T4, no install needed. Combines FloodDiffusion motion generation with a LLM cognitive brain (Perceive → Predict → Decide → Act). Scored 726/1000 (Grade B) on Track C — the only directly verified model so far. Submissions from other teams very welcome.
---
🗂 Dataset → FINAL-Bench/World-Model
🌍 Demo → FINAL-Bench/World-Model
🏆 Leaderboard → FINAL-Bench/worldmodel-bench
📝 Article → https://huggingface.co/blog/FINAL-Bench/world-model
Part of the FINAL Bench Family — alongside FINAL Bench (Feb 2026). Feedback on rubrics and missing models always welcome!
upvoted a paper 5 months ago