- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper introduces MuSEAgent, a framework for improving multimodal tool-using agents through reusable, state-conditioned experiences. Offline trajectories are decomposed into atomic transitions, and a hindsight model produces a score and concise natural-language guidance for each transition. The framework filters experiences using graph-smoothed scores and indexes each state through 15 combinations of its instruction, visual input, task descriptor, and execution history. At inference time, Deep-and-Wide retrieval aggregates guidance across multiple queries before generating the next action.
Experiments on four high-resolution visual question-answering benchmarks show improvements over ReAct, Reflexion, ExpeL, and Memento across three Qwen-based agents. The paper also investigates cross-dataset transfer, experience sources, and alternative hindsight models.
Critical strengths
-
The problem and motivation are clear, and the novelty is reasonable. Experience itself does present the problem of being long and containing useless information. Atom transition can actually make it more concise and more useful for generating high-quality experience.
-
The reported experiment results show improvement across models and datasets.
Critical weaknesses
-
The experiments are mostly done on Qwen series models. And more importantly, I could not understand why the authors mainly do experiments on multiple-choice benchmarks. This should do experiments on more general tasks, not multiple choice such easy tasks.
-
The claimed iterative refinement is not established by the algorithm or the comparisons. In Algorithm 2, every viewpoint is sampled conditional on the same state. Previously retrieved guidance does not explicitly enter subsequent viewpoint selection, and the aggregated guidance is consumed only after the retrieval loop.
Questions for authors
-
Can you do more experiments on general evaluation benchmark and other series models?
-
Does the score represent action quality, error responsibility, or the expected usefulness of the resulting guidance? How are successful and failed transitions made comparable before smoothing and filtering?
-
Does each retrieval round observe the results of previous rounds? Please provide an execution trace showing viewpoint selection, retrieval, and action generation, including model-call counts.