- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
REDUCE is a training-free two-level context compression framework for terminal agents. It filters redundant terminal output via an updatable rule pool, then compresses history through fact-verified structured summaries. It designs a risk-aware controller to dynamically adjust filtering, retention windows, and compression thresholds based on token pressure, verification results, and rule feedback.
Critical strengths
- REDUCE combines observation filtering and history compression, dynamically adjusting compression intensity based on shared risk signals rather than fixed thresholds.
- Summary validation, deferred retries, and extractive fallback help prevent unreliable summaries from replacing important historical information, improving the robustness of context compression.
Critical weaknesses
- Limited comparison with related methods. REDUCE does not include a direct comparison with ACON, which makes it difficult to assess the benefits of the proposed risk-aware coordination.
- Insufficient ablation of key components. The paper does not isolate the contributions of grounded validation, deferred retries, extractive fallback, or adaptive control.
- Limited analysis of compression risks. The experiments mainly report task accuracy and token consumption. A case study of compression risks will be beneficial.
- Limited evaluation of efficiency and robustness. The paper reports substantial token savings but does not evaluate end-to-end latency or actual inference costs.
- Insufficient experiments. Only three models are evaluated on three datasets, and on CompileBench and DevEval only the results of MiniMax-M3 are reported, with no comparison provided.
- Format issues. Figure 3 is hard to read and understand.
Questions for authors
- Can the token savings transfer into lower end-to-end latency and inference costs? How much overhead is introduced by summarization, validation, and rule evolution?
- REDUCE has many heuristic designs, and the actual effectiveness of each design remains unclear. Moreover, the performance of these designs varies across benchmarks and models, further limiting the method's generalizability.