- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
UPTURN corrects reasoning trajectories by selecting a paragraph boundary, retaining the prefix, inserting a fixed instruction as a new user turn, and regenerating the suffix. It trains lightweight position-selection and gating heads while keeping the reasoning model and encoder frozen. Experiments cover multiple mathematical benchmarks, cross-model transfer, and generalization to GPQA-Diamond.
Critical strengths
-
Lightweight and practical design. UPTURN learns intervention positions from final-answer correctness without step-level annotations or full-model fine-tuning. This makes it a potentially useful addition to existing reasoning models.
-
Substantial tight-budget improvements with broad experimental support. Results across multiple models and benchmarks, together with injection, position-selection, gating, and seed ablations, support the approach. End-to-end runtime measurements further strengthen its practical relevance.
Critical weaknesses
-
The advantage diminishes substantially as the generation budget increases. At 24K, Avg gains over Base@32 are +0.6 percentage points for Qwen3-4B, −1.6 for Qwen3-8B, and +7.0 for DeepSeek-R1-Qwen3-8B, compared with +14.1, +16.9, and +30.3 at 8K. Both methods generally improve in absolute accuracy, but independent sampling catches up. It therefore remains unclear how much of UPTURN’s benefit comes from recovering length-truncated trajectories versus correcting completed but incorrect solutions. Reusing the 8K-trained position policy also complicates the interpretation of larger-budget results.
-
Comparative evaluation and reporting need strengthening. Closely related rollback-based approaches such as TAAR, Thought-ICS, and StepCo warrant discussion and representative comparisons. The manuscript and supplementary main-table code also differ in gate threshold and position selection, limiting reproducibility. As a presentation improvement, Figure 1 should be redrawn: its decorative design obscures the mechanism. A simpler side-by-side diagram should clearly identify the cut position, retained prefix, turn boundary, and regenerated suffix for inline versus user-turn injection.
TAAR https://aclanthology.org/2026.findings-acl.1930
Thought-ICS https://arxiv.org/abs/2602.02416
Questions for authors
-
Could you report length-truncation rates at each budget, together with correction success rates separately for truncated trajectories and completed but incorrect trajectories? These statistics would help explain the different outcomes at 24K.
-
Could you extend evaluation to at least 32K, and preferably 48K or 64K where supported, reporting accuracy, uncertainty, actual generated tokens, and runtime? Comparing the reused 8K policy with a budget-specific policy at one larger budget would help isolate transfer effects.
-
Please clarify whether the reported results use the manuscript’s 0.4 gate threshold and sampled positions or the supplementary script’s 1.0 threshold and
argmaxselection, and provide matching evaluation commands. A representative comparison with one of the closely related methods above would also strengthen the assessment.
2 comments
R3. Gate threshold and position selection (W2, Q3)
Gate threshold. To avoid repeating generation across analyses, we intentionally use γ=1.0 to precompute one correction for every rollout. We subsequently recomputed the metrics from the saved rollout dumps, which contain the original and corrected responses and their gate scores. For the reported main results, we apply γ=0.4 to the original rollout scores and replace only responses below this threshold.
Position selection. Thank you for pointing this out. The sampling description was our mistake: training samples G=8 cuts without replacement, while inference uses argmax. We have corrected Sec. 5.3 and Appendix C. Revisions are highlighted in blue.
R4. Comparison with related rollback methods (W2)
We have added TAAR, Thought-ICS, and StepCo to Related Work, with the additions highlighted in blue.
TAAR [1] learns trap locations and escape probabilities to control truncation and restart with a frozen reasoner. Our reduced reimplementation achieves 44.14% Avg / 57.61% Pass, compared with UPTURN's 56.2% / 73.4% (six-benchmark means; separate rollout pools). Keeping TAAR's cuts and English instruction fixed, moving the instruction into a new user turn improves performance to 48.19% / 67.46%. The full comparison is in KVSn R1.
Thought-ICS [2] structures generation into discrete thoughts, verifies the response, localizes the first erroneous thought, and iteratively backtracks to resample from the preceding prefix. Its explicit error-localization loop differs from UPTURN's policy trained on regeneration outcomes from existing responses.
StepCo [3] alternates verification by a process-supervised verifier with revision of incorrect reasoning steps. UPTURN learns where to cut from final-answer rewards and introduces a new user turn for suffix regeneration, without requiring step-level error labels for policy training.
Our contribution combines outcome-trained cut selection with user-turn injection, supported by the TAAR comparison and our inline/user-turn ablation.
R5. Figure 1 revision (W2)
We have updated Figure 1 in the revised paper with a simpler side-by-side comparison of inline and user-turn injection, explicitly labeling the cut position, retained prefix, turn boundary, and regenerated suffix. The caption has also been updated, with revised text highlighted in blue.
[1] Chen, Kang, et al. "Thinking Traps in Long Chain-of-Thought: A Measurable Study and Trap-Aware Adaptive Restart." arXiv preprint arXiv:2601.11940 (2026).
[2] Samanta, Ankur, et al. "Structure enables effective self-localization of errors in LLMs." arXiv preprint arXiv:2602.02416 (2026).
[3] Wu, Zhenyu, et al. "Enhancing mathematical reasoning in llms by stepwise correction." Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
Please let us know if you have any further questions or would like additional clarification.
Thank you for your thoughtful comments and constructive suggestions. We address your concerns below.
R1. Budget scaling and correction success (W1, Q1)
We report equal-weight averages over six benchmarks and Qwen3-4B/8B and DeepSeek-R1-Qwen3-8B, using our learned gate (γ=0.4). T→C / selected T and W→C / selected W measure correction success among selected truncated and wrong responses, respectively. Undefined transition rates are excluded from averaging.
The declining gains cannot be explained solely by the increasing share of wrong responses: T→C success also decreases with the budget, and wrong responses remain less frequent than truncated responses through 24K. Moreover, the SCoRe experiment below shows that UPTURN can correct wrong answers. These findings suggest that the decline may reflect limitations in transferring the 8K-trained policy to longer rollouts; the comparison in R2 examines this possibility.
To evaluate wrong-answer correction with an 8K initial-generation budget, we examine the response pools from our complementary experiments, under the same averaging and gating settings:
SCoRe produces more wrong than truncated responses. Applying the existing UPTURN policy with “Resolve this concisely.” (γ=0.4), without additional training, gives:
C, W, and T denote correct answers, wrong extracted answers, and responses with no extracted answer, respectively. W→C gain and T→C gain are the accuracy gains from correcting W and T responses; C→W/T loss is the accuracy lost when initially correct responses become wrong or yield no extracted answer. Net gain is the sum of the two gains minus this loss. The contributions below are percentage points across all responses, rather than the conditional success rates above:
Wrong-answer correction contributes more to this gain than recovery of truncated responses, showing that UPTURN also turns wrong extracted answers into correct ones.
R2. 32K evaluation and budget-specific policy comparison (Q2)
The requested 48K/64K response budgets exceed Qwen3-4B’s configured 40,960-token context window, which includes both the prompt and response. We therefore extended evaluation to 32K on three reasoners across all six benchmarks. Accuracy denotes Avg., with equal weights across benchmarks and reasoners. We use γ=0.1, selected from a post-hoc threshold sweep.
UPTURN improves accuracy by 1.35 pp over Base@16, the same 16 initial responses without correction (95% CI [0.71, 2.07]). To compute the 95% CIs, we use 10,000 bootstrap resamples within each benchmark, sharing the sampled questions across reasoners and methods. These CIs reflect variation in evaluation questions only; models, generated responses, and γ are held fixed.
Representative computation cost. We measured Qwen3-4B on 30 AIME24 questions at 32K using one H200 (TP=1). Base@32 samples 32 responses in one batch; Base@16 measures the generation of UPTURN’s 16 initial responses, and UPTURN@16 additionally applies correction using the 8K-trained policy. Results are averaged over two runs; runtime is reported as mean ± SD.
UPTURN uses 43.5% fewer generated tokens and 34.8% less runtime than Base@32. Token counts include newly generated initial responses and correction suffixes, excluding retained prefixes.
At 16K evaluation, we compare 8K-trained and 16K-trained policies with γ=0.4. Scores are equal-weight averages over six benchmarks and three reasoners.
The 16K-trained policy achieves higher mean accuracy than the 8K-trained policy (67.1% vs. 66.6%, a gain of 0.5 pp).