acceptodds
Under review as a conference paper at ICLR 2027

Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts

Abstract

Research agents explain planned experiments. We measure predictive credit with paired forecasts sharing an intervention, forecaster, and outcome while varying description, matched explanation, and donor context. Five checks track commitment, delivery, predictive gain, alignment, and known-signal uptake. Across 336 prospective states in controlled learning, 12 Tox21 endpoints, and 24 OpenML tasks, v5's frozen credit decision was inconclusive. Tox21's preregistered ROC AUC interval-score harm test was unmet (, 95 percent interval [, .0104]); OpenML's joint formation, point-equivalence, and repeatability rule was unmet. Matched point-accuracy gains over description remained unconfirmed, and Tox21/OpenML seed-donor intervals spanned zero. Under requested DeepSeek V4 Pro, matched and donor cards reduced secondary Tox21 drift by 64.5 and 59.1 percent. A DeepSeek V4 Flash replay raised matched point MAE from .01823 to .02020 and missed matched-donor interval-score equivalence. OpenML full-card assignment widened nominal 80 percent intervals by 21 percent, with 49.3 percent coverage versus 51.4 percent for description and content in 66/144 cards. Direct-text Flash delivered all 144 notes without detectable matched point-accuracy gain. A researcher-authored mechanism positive control lowered point MAE by 2.60 percentage points versus description. The protocol measures predictive credit for research-agent benchmarks and scientific forecasting; natural-explanation credit remained unconfirmed at the tested donor resolutions.

Primary area:
interpretability and explainable AI
Keywords:
AI research agents, scientific explanations, experimental outcome prediction, predictive evaluation, automated scientific discovery, uncertainty calibration, large language models

Reviews

Among:
Shape:
1234this paper, Bottom 1%
Figure 1. Mean review rating of the reviewed papers at ICLR 2027. This paper averages 1.00 over 3 reviews, above 0% of the others.
Reviewer_yryEOpenReview ↗
Rating
1
Confidence
3
Informativeness
1
Presentation
1
Soundness
1
Read review
Paper summary

The authors study whether LLM-generated explanations of planned ML experiments help an LLM forecaster predict the experiments’ outcomes. For each experiment, the forecaster (DeepSeek V4) predicts the outcome under three contexts, twice each: from the description alone, with the matching explanation, and with an explanation borrowed from another experiment. The explanations are generated by single, tool-free DeepSeek V4 calls, mostly for pre-assigned hyperparameter or feature changes.

The evaluation covers 336 small experiments across synthetic tasks, 12 Tox21 assays, and 24 OpenML datasets with LightGBM. It adds several post-hoc follow-ups and two positive controls.

None of the three main tests passes under the decision rules the authors fixed before seeing the outcomes. Explanations do not measurably improve accuracy. In Tox21 and OpenML, matched and borrowed explanations give similar results; only in v5 do matched explanations beat borrowed ones, on log-loss error. Repeated forecasts become closer to each other in Tox21, but not on OpenML. The authors present the paired comparison design itself as the main contribution: a protocol for evaluating research-agent explanations.

Critical strengths

The study fixes its decision rules before seeing outcomes, keeps failed calls in the analysis, treats tasks as the unit of inference, and includes positive controls showing the forecaster can use real information. The negative results are reported. These are good practices, but they do not offset the weaknesses below.

Critical weaknesses
  1. The paper is framed around research agents, but no agent is evaluated. Each explanation is a single, tool-free DeepSeek V4 call that sees only the public state and the intervention, with no code execution and no interaction with the experiment. The interventions themselves are assigned by rule (Tox21, OpenML) or chosen from a fixed catalog (v5). The results therefore concern single-call LLM rationales for pre-specified configuration changes; applicability to research-agent benchmarks (§8–9) is asserted, not tested.

  2. The drift reduction comes from the number on the card, not from the explanation. In the prospective studies, the explanation and the forecast come from the same model with the same inputs (lines 146-151, 265-267), so the card adds no information, only a predicted number (lines 182-183). Forecasts that see the same number agree with each other without getting closer to the truth. The authors note this (lines 190-192), yet the abstract and Figure 1 highlight the 64.5%/59.1% drift reductions.

  • Numbers alone cut drift more than mechanism text (Table 25: 0.00246 vs 0.00378). With the mechanism text held fixed, numbers from the borrowed explanation cut drift at least as much as the matched ones (0.00256 vs 0.00317). The paper labels this contrast exploratory, but it is consistent with the number, not the explanation, driving the effect.

  • Point MAE does not meaningfully improve (Table 2) and worsens in the Flash replay, from 0.01823 to 0.02020.

  • Real information raises drift while lowering error (Table 28: drift goes from 0.768 to 1.275).

  • On OpenML, where most cards have no number, drift does not fall (0.01004 under description, 0.01300 with cards, line 322).

  • The reduction is unstable across identical Flash runs: 0.00426 in one, 0.00108 in the other (lines 314-316).

  1. The forecaster is no better than always guessing "no change". The simplest forecast is to predict that the score will not change at all. The authors use this as a baseline (lines 199-200), but their forecaster does not beat it. On OpenML, its error is the same as the baseline (0.08368 vs 0.08341, lines 435-437), and it does better on only 13 of 24 tasks (Table 22). It only wins after one dataset, brazilian_houses, is removed (lines 436-437). For Tox21 and v5, the comparison is never shown. Without evidence that the forecaster extracts signal from the description alone, it is unclear whether these tasks contain enough predictable signal for an explanation to help, which limits what the null results can show.

  2. The tests cannot detect small differences. In Tox21 and OpenML, the borrowed explanation comes from the same experiment run with a different random seed (lines 142-143), so its true outcome is usually close to the matched one. The two seeds usually move in the same direction (27 of 36 and 54 of 72 pairs), and in Tox21 the gap between them (0.0136) is smaller than the forecaster’s own error (about 0.02) (lines 143-144). A test like this cannot tell whether an explanation is specific to its own experiment. On OpenML there is almost nothing to compare. Only 13 of 72 pairs had content in both explanations, and only 1 pair had numbers in both (lines 423-424). The sample is small. Each task counts as one unit (lines 288-289), so Tox21 rests on 12 tasks and OpenML on 24. The thresholds are not justified. Two results are called equivalent if they differ by less than $\pm$0.002, $\pm$0.005 or $\pm$0.01 (Table 26; lines 1056, 1061), but the authors never say where these values come from.

Questions for authors
  • Other models: would the results hold with a model family other than DeepSeek, or with higher reasoning effort? Most calls used low effort (line 268).

  • 78 of 144 cards were empty (line 423), and extraction failed your own .95 quality threshold (lines 1150-1151). What can the OpenML comparison tell us when most forecasts receive no explanation?

  • The positive control uses a researcher-written note on synthetic tasks. Would explanations written by domain experts help the forecaster on Tox21 or OpenML? That would show whether the problem is the explanations or the method.

Reviewer_mVqYOpenReview ↗
Rating
1
Confidence
2
Informativeness
1
Presentation
1
Soundness
1
Read review
Paper summary

The paper appears to study whether an agent’s explanation of a planned experiment helps predict its outcome. However, I could not fully understand the proposed method or identify the precise new contribution from the current presentation.

Critical strengths

None

Critical weaknesses

The presentation of this work is poor and difficult to follow. For instance, the paper mentions labels/concepts such as “v5,” “frozen credit decision,” and “confirmation no go” without explaining them. The submission needs substantial rewriting and restructuring for a clear, self-contained introduction, problem statement, method, and central result.

Questions for authors

None

Reviewer_KwgKOpenReview ↗
Rating
1
Confidence
3
Informativeness
1
Presentation
1
Soundness
1
Read review
Paper summary

The paper introduces a method to measure “predictive credit” for LLM agents in scientific experiments. The main goal is to analyze whether providing an agent with an explanation that motivates a given experiment can improve the accuracy in forecasting the experiment’s outcome. The authors test this with two models, DeepSeek V4 Pro and DeepSeek V4 Flash, across 3 settings. They do not find strong evidence that explanations improve forecasting accuracy.

Critical strengths

The main topic of the paper, AI-driven scientific discovery, is important and timely.

Critical weaknesses

The paper contains major methodological issues. In OpenML experiments (one of the three experimental settings), more than half of extracted explanation cards provided to the model were empty (Table 18, empty extracted cards make up 54%). This seriously undermines any conclusions drawn from the main OpenML experiment on whether an explanation improves the model’s forecasting accuracy. The experiments also rely only on two models from the same family, limiting the generalizability of the results.

The authors do not clearly define several important terms that are used throughout the paper, such as “confirmation_no_go”. This makes it very difficult to understand what conclusions the authors draw from their own results, and whether the experiments sometimes fail the authors’ own internal tests. Overall, the paper is very difficult to follow.

Questions for authors

Can you clarify the meaning of the “confirmation_no_go” term?

From OpenReview, CC BY 4.0. Authors as signed there; shown unedited.

Related papers

Loading the map…

Discussion (0)

Sign in to comment.