- Rating
- 3(+1)
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 1
- Soundness
- 2(+1)
Read review
Paper summary
Large language models (LLMs) can optimize the solution recursively based on the feedback provided by the domain-specific verifier, forming a self-evolution loop. However, some domains cannot obtain high-fidelity verifiers and instead rely on low-cost but imperfect proxy rewards. To improve the quality of the verifier and therefore the performance of the self-evolution outcome, this paper proposes Conformal Interval-Driven Self-Evolution (CISE). Specifically, CISE replaces single-point proxy rewards with calibrated property intervals and guides material search using conservative interval feedback and full-interval compliance rules. Three material tasks show that DFT false positives among the returned candidates are significantly reduced. The method has clear value in addressing the issue of self-evolution under biased proxy feedback.
Critical strengths
- The paper focuses on an important problem: that scientific agents may repeatedly reinforce wrong candidates when chasing high proxy scores.
- Extensive experiments have been conducted to provide a detailed discussion of the components of the proposed method.
Critical weaknesses
- The conference seems not to allow directly showing the GitHub link in the submission.
- The paper proposes a better reward for the evolving tasks. However, it remains unclear about the significance of the tasks. For example, how could an imperfect proxy reward infer the evolution process and final results?
- Theorem 2 requires the calibration samples to remain i.i.d. given the generated state, but the states come from previous intervals and archive updates using the same calibration set.
- The experiment measures the feasibility of the final candidate and has not yet verified the coverage per round and the width of the online interval.
Questions for authors
- Please discuss or provide results on how an imperfect proxy reward could infer the evolution process and final results.
- Please explain how the theorem applies to actual closed loops.
8 comments
[W2, Q1] How imperfect proxy rewards affect evolution and final outcomes
The issue is that a proxy error can affect both the current selection decision and the evidence used to generate later candidates. A candidate may satisfy even though . Reusing its favorable point prediction as evidence of success can then guide subsequent proposals toward the same misleading feature combinations. The motivation is therefore not simply to change the final filter, but to improve the reward information used throughout search.
The revised Figure 1, Algorithm 1, and Section 3.3 show this feedback path explicit. CISE uses historical inspection data to initialize a hypothesis about feature combinations associated with successful rewards and short intervals. The hypothesis guides candidate generation and determines the calibration space . After evaluation, both accepted and rejected candidates are stored in with their features and interval rewards. This evidence informs the next hypothesis, while only candidates satisfying enter the output set. The proxy and LLM parameters remain fixed, and newly proposed candidates' true rewards are not used during search.
We have added GB1 protein sequence optimization to examine this process beyond materials discovery. The revised Figure 2 displays proxy predictions and interval rewards alongside experimentally measured fitness, with candidates grouped by their generating hypotheses. Several early hypotheses yield few accepted candidates, whereas later stages, particularly , generate many true positives. This trajectory provides evidence about how interval-informed hypothesis revision accompanies successful discovery, rather than only reporting the final output set.
The quantitative evidence also considers discovery yield, not just rejection of false positives. Under 400 candidate sequences per method on GB1, CISE returns 60 TP and 18 FP, with precision 0.77 and best fitness 4.470; LEON returns 59 TP and 106 FP, with precision 0.36 and best fitness 3.876 (revised Table 1). The component comparison in the revised Table 3 further shows:
| Method | WBG Prec. | WBG TP | SSE Prec. | SSE TP |
|---|---|---|---|---|
| Proxy | 0.55 | 63 | 0.54 | 70 |
| Proxy + CCI | 0.97 | 60 | 0.91 | 48 |
| + Hypothesis (CISE) | 0.93 | 96 | 0.96 | 136 |
The gain from Proxy + CCI to full CISE supports the role of the hypothesis-driven component, which couples generation with its calibration basis, in expanding successful discovery while retaining high precision.
[W3, Q2] Scope of the coverage result under adaptive calibration-data reuse
You are correct that conditioning on a generation state obtained from earlier interval feedback does not, by itself, preserve the required independence of the calibration samples. We have revised the theoretical statement and its interpretation to separate the i.i.d. reference setting from the experiments.
Nevertheless, Theorem 1 provides the i.i.d. reference for CCI interval construction; our central objective is self-evolution driven by calibrated interval rewards that improves both output precision and TP count. Its group-conditional guarantee motivates robustness to group-marginal shift: changes in group proportions with unchanged within-group distributions, under the reference independence assumptions.
CISE connects this principle to generation by restricting hypotheses to over the fixed feature map . The feature conditions in both guide candidate generation and define the calibration groups represented in (Section 3.3). Refinements require executable predicates, sufficient inspection-data support, and a limited number of splits. Thus, generation and calibration use the same feature-defined groups, aiming to align search-induced shifts with distinctions captured by the calibration basis.
Section 4.1 distinguishes this reference guarantee from adaptive fixed-data reuse. The practical contribution is supported directly by Table 3: adding hypothesis-driven evolution to Proxy + CCI increases TP count from 60 to 96 on WBG and 48 to 136 on SSE, with final precision of 0.93 and 0.96, respectively. The new Figure 2 further shows productive later hypotheses on GB1, supporting interval feedback as a mechanism for guiding subsequent generation, rather than only filtering final outputs.
[W1] Anonymous code repositories are permitted by the conference policy
Thank you for raising this policy question. The ICLR 2027 Author Guideline* explicitly permit anonymous repository links. In its FAQ on anonymous code sharing, the guidelines state:
Thus, an anonymous repository link is an allowed way to share code; the relevant requirement is preserving anonymity, rather than a blanket prohibition on GitHub links. Our inclusion of the repository link was intended to support reproducibility under this option.
https://iclr.cc/Conferences/2027/AuthorGuidelines.