The authors introduce MMPostTrainBench, covering eight multimodal post-training tasks.
Research agents struggle to sustain gains in multimodal model training
Across eight tasks, autonomous post-training frequently degraded models, while a framework using evidence, memory and candidate selection improved final submissions.
Chinese Tech
Yuxin Liu · Yuxuan Wang · Zhenxin Lei · Lingchen Meng · Yuchong Sun · Junming Lin · +6 more
University of Science and Technology of China · Alibaba Token Hub, Alibaba Group · University of the Chinese Academy of Sciences · Tsinghua University · Shanghai Jiao Tong University
Research Digest··3 min read
Liu and colleagues evaluate whether research agents can autonomously improve a common base model on image, audio, video, joint audio-video and image-grounded software tasks.
Why this paper
From University of the Chinese Academy of Sciences and 4 others
In one line
Autonomous research agents for multimodal post-training often degrade performance and fail to sustain improvements.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§