- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper studies how to improve multimodal reasoning while compressing explicit reasoning traces into short latent trajectories. It proposes FMLR, which supervises intermediate latent states using joint representations of reasoning text and corresponding visual evidence, together with MSL-GRPO to optimize latent reasoning, answer generation, and stopping through correctness, early-sufficiency, and length rewards. The authors also construct refined multimodal reasoning traces and corresponding embedding datasets. Evaluation includes eight vision-centric benchmarks, data-matched supervision and optimization ablations, an accuracy–output-length analysis, and experiments on two additional backbones. The reported results show accuracy gains for the complete training recipe and substantially shorter outputs than the Thinking and Refine CoT baselines.
Critical strengths
-
This paper introduces a framework that supervises latent reasoning through joint embeddings of visual evidence and textual explanations, while incorporating reinforcement learning to optimize both reasoning and termination decisions. The approach strengthens visual reasoning without requiring extensive textual chains of thought. With the backbone and training data held constant, the full method achieves higher accuracy across all eight benchmarks, exceeding MSL with independent supervision by 0.5 to 7.7 percentage points and standard GRPO with fused supervision by 0.5 to 4.8 percentage points. On the four benchmarks reported in Table 3, it also outperforms Refine CoT by 1.6 to 5.0 percentage points and reduces the reported output length to approximately 9 to 10 tokens, compared with 156 to 217 tokens for Refine CoT. These results demonstrate its ability to deliver higher accuracy alongside substantially more concise outputs.
-
This work develops reusable intermediate supervision by refining generated reasoning trajectories and aligning textual explanations with relevant image regions. The benefits of this refinement extend to explicit reasoning independently of the specific latent architecture used. Refine CoT achieves better results than Auto-CoT on seven of the eight benchmarks, with improvements of 9.4 percentage points on MME-RW and 9.9 percentage points . The resulting fused targets can also be reused across different backbones without requiring them to be embedded again. Using these targets, the full training method improves upon the corresponding base models on all eight benchmarks for InternVL3.5-8B and seven of the eight benchmarks for Qwen3-VL-4B. These findings highlight the practical value of constructing supervision that remains useful beyond the primary model.
Critical weaknesses
-
The paper does not establish whether the early-sufficiency reward and learned stopping reduce latent computation without sacrificing accuracy. Table 2 reports accuracy gains, and Table 3 demonstrates compression relative to explicit CoT, but neither compares the actual latent-step counts of fused SFT, standard GRPO, and MSL-GRPO. Starting from the same SFT checkpoint and using the same latent/stop/answer policy, the authors should compare correctness-only reward, correctness plus a length penalty, and the full MSL reward, reporting both accuracy and average latent steps. Comparing learned stopping against fixed budgets of 2, 4, and 8 steps would further establish whether learning when to stop offers an advantage. These tests are needed to substantiate the specific benefits claimed for MSL-GRPO beyond the effectiveness of the complete training recipe.
-
The comparison with prior latent reasoning methods leaves a consequential uncertainty about training conditions. Table 1 states that LVR-8B and Monet-8B use the same backbone as FMLR, but does not specify whether they receive the same refined training data or comparable training and inference budgets. This matters because Table 2 itself shows that data refinement substantially changes performance—for example, MME-RW rises from 42.0 to 51.4 under explicit CoT supervision. The authors should disclose the baseline configurations and, if these conditions are not already matched, compare FMLR with Monet using the same refined examples and comparable budgets. Although the internal ablations support FMLR’s effectiveness, this comparison is needed to determine how much its advantage over the closest prior method comes from the proposed method rather than differences in supervision or computation.
Questions for authors
-
Could the authors clarify exactly how “Fused Latent + GRPO” differs from MSL-GRPO in Table 2? Does it optimize the same latent actions, stopping decisions, and answer tokens under the same rollout budget, and which reward terms does it use? If relevant ablations are already available, please provide accuracy and average latent-step counts with and without the early-sufficiency bonus and length penalty, together with any comparison of learned versus fixed-length stopping. These results would help establish whether the proposed reward and stopping mechanisms provide benefits beyond jointly optimizing latent reasoning and answer generation.
-
Were the LVR-8B and Monet-8B baselines trained on the same refined Visual CoT examples and ZebraCoT subset as FMLR? Please clarify any differences in training budgets and inference settings, particularly the number of latent steps. If these conditions differ, are comparisons under matched conditions already available? This clarification would affect my confidence that the improvements over prior methods come from the proposed method rather than differences in training data or computation.
4 comments
2. W2 / Q2 — Shared supervision and training/inference budgets for LVR and Monet
Thank you for emphasizing the effect of data refinement. The LVR-8B and Monet-8B entries are our reimplementations on the same Qwen3-VL-8B backbone, refined Visual CoT examples, and ZebraCoT subset as FMLR. We now state this configuration explicitly and disclose their training and inference budgets.
Same refined supervision. Table 2 shows the value of refinement: explicit Auto-CoT and Refine CoT achieve 42.0% and 51.4% on MME-RW, respectively. All three latent methods use the refined data. Stage 1 contains 403,593 examples; Stage 2 uses the ZebraCoT subset.
Complete eight-benchmark results from Table 1. Accuracy (%) ↑:
| Benchmark | N | LVR-8B ↑ | Monet-8B ↑ | FMLR-8B ↑ |
|---|---|---|---|---|
| MMVP | 300 | 77.7 | 78.3 | 80.3 |
| HR-4K | 800 | 70.0 | 76.2 | 78.8 |
| HR-8K | 800 | 63.8 | 70.0 | 71.9 |
| AI2D | 3,088 | 81.7 | 82.3 | 83.3 |
| MME-RW Lite | 1,919 | 50.3 | 53.6 | 58.4 |
| V* | 191 | 82.7 | 83.2 | 84.3 |
| TextVQA | 5,000 | 80.6 | 80.1 | 81.9 |
| VisualPuzzles | 1,168 | 34.4 | 36.9 | 37.2 |
FMLR improves over both reimplemented methods on all eight benchmarks. Against Monet, the gains include 2.0 points on MMVP, 2.6 on HR-4K, and 4.8 on MME-RW; the MME-RW gain over LVR is 8.1 points.
Inference budgets. LVR-8B and Monet-8B each execute eight latent steps. FMLR learns to stop within one to eight steps. Table 4 uses LVR-7B for the timing comparison. Table 5 compares learned and fixed 2/4/8-step inference within FMLR.
SFT budget and measured time. All methods use 403,593 examples. FMLR completes one epoch in 1,577 updates; LVR uses the same sample count; Monet follows its three-stage schedule. Table 16 records wall-clock hours on NVIDIA A800 80GB GPUs:
| Model | Stage 1 h ↓ | Stage 2 h ↓ | Stage 3 h ↓ | Total h ↓ |
|---|---|---|---|---|
| FMLR-8B | 34.83 | — | — | 34.83 |
| LVR-8B | 42.1 | — | — | 42.1 |
| Monet-8B | 104.5 | 66.3 | 47.5 | 218.3 |
RL exposure and measured time. The common one-epoch budget contains 5,872 prompts and four trajectories per prompt. Batch sizes lead to different update counts, which Table 17 reports explicitly:
| Model | Prompts | Rollouts/prompt | Updates | RL h ↓ |
|---|---|---|---|---|
| FMLR-8B | 5,872 | 4 | 734 | 3.12 |
| LVR-8B | 5,872 | 4 | 734 | 13.0 |
| Monet-8B (VLPO) | 5,872 | 4 | 92 | 18.3 |
SFT and RL timing exclude offline target precomputation, evaluation, and judging. FMLR takes 34.83 h for SFT and 3.12 h for RL under these configurations, with the highest accuracy across all eight benchmarks.
Manuscript changes. §4.1, p. 7 specifies the common backbone, refined examples, ZebraCoT subset, and eight-step baseline inference. Table 1, p. 8 provides all benchmark scores; Appendix G, Tables 16–17, p. 34 gives stage-wise SFT and RL budgets. The revised manuscript is provided in the Supplementary Material with changes highlighted in blue.
Thank you for the additional experiments and clarifications. I still have two questions.
1. Does learned stopping actually achieve adaptive reasoning?
Both the Full reward configuration and the learned stopping experiment report an average of 1.000 latent steps. Does this mean the model always stops after the first step? If so, it seems to behave like a fixed one-step strategy rather than an adaptive stopping policy. Could the authors provide the distribution of stopping steps and results for a Fixed-1 baseline?
2. What is the main benefit of the learned stopping mechanism?
C+L achieves slightly higher average accuracy than Full (72.066% vs. 72.013%), while Full uses fewer latent steps. Also, learned stopping has nearly the same latency as Fixed-2 (2.706s vs. 2.710s). Could the authors clarify whether the main benefit is accuracy or efficiency? A comparison under matched latent-step budgets would help establish the actual contribution of learned stopping.
Thank you for these follow-up questions. Our objective is straightforward: help the model answer correctly as early as possible, while avoiding unnecessary latent computation.
1. Learned stopping and effective one-step reasoning
MSL-GRPO jointly rewards answer correctness and early sufficiency while discouraging unnecessary latent steps. The model can use between one and eight steps and learns when to stop. On the reported 700-question panel, it consistently chooses to answer after its first latent step: 700 questions use one step (100%), and none use two to eight steps.
The reported comparisons show that this is an effective operating point. Learned stopping achieves 70.429% accuracy, exceeding Fixed-2 by 1.71 percentage points and both Fixed-4 and Fixed-8 by 2.00 percentage points. What matters is how effectively the model uses its latent reasoning, rather than the number of steps alone.
We further compare the reward configurations when all conditions execute one latent step on the same 700 questions:
Table 1. Accuracy of MSL-GRPO reward configurations at a matched one-step budget on 700 questions.
| Reward configuration | Stopping rule | Correct / Total | Accuracy |
|---|---|---|---|
| C | Fixed-1 | 469/700 | 67.000% |
| C+L | Fixed-1 | 486/700 | 69.429% |
| C+S | Fixed-1 | 482/700 | 68.857% |
| Full | Learned stopping (one step on every question) | 493/700 | 70.429% |
Full achieves the highest accuracy among these four configurations at the same one-step budget, outperforming C, C+L, and C+S by 3.43, 1.00, and 1.57 percentage points, respectively. This shows that the complete reward design produces a more effective one-step answering policy.
The comparison with non-latent reasoning provides complementary evidence. In Table 2, Full outperforms Non-latent CoT + GRPO on all eight benchmarks, increasing macro accuracy from 67.538% to 72.013%, an improvement of 4.475 percentage points.
Together, these comparisons highlight the value of learning to use latent computation effectively. The policy learns its stopping decision within the available budget; on this panel, that learned decision is consistently one step. The resulting policy achieves strong performance with a compact trajectory and benefits from the complete reward design even when the compared configurations use the same number of latent steps.
2. Accuracy and efficiency
Table 1 shows how effectively each reward configuration uses a single latent step. With all three MSL-GRPO reward components, Full learns to stop at the earliest permitted step on all 700 questions. It achieves the highest accuracy among the four configurations at this matched budget (70.429%), exceeding C, C+L, and C+S by 3.43, 1.00, and 1.57 percentage points, respectively.
Across the eight benchmarks, Full maintains nearly the same macro accuracy as C+L (72.013% versus 72.066%) while using 41.2% fewer latent steps on average (1.00 versus 1.70). On the 700-question panel, learned stopping and Fixed-2 have nearly identical mean latency (2.706 s versus 2.710 s), while learned stopping achieves 1.71 percentage points higher accuracy with half as many latent steps. Compared with Fixed-8, it improves accuracy by 2.00 percentage points and reduces mean latency by 10.8%.
These 8B results motivate exploring compressed reasoning at larger scales. For a dense 100B decoder, saving one decoding step per answer would avoid roughly 200 billion floating-point operations in matrix multiplications alone. This estimate illustrates the potential value of preserving accuracy with shorter reasoning sequences across many queries. This makes preserving accuracy while reducing the total number of decoding steps a promising direction for future work. Pope et al., Efficiently Scaling Transformer Inference, MLSys 2023
1. W1 / Q1 — Correctness-only GRPO, MSL rewards, and learned stopping
Thank you for the suggestion. We now report the correctness-only baseline, all reward combinations, actual latent counts, and fixed-budget inference results.
What changes from GRPO to MSL-GRPO. The original “Fused Latent + GRPO” row is correctness-only C, now labeled in Table 2. C, C+L, C+S, and Full start from the same fused-SFT checkpoint. All optimize latent actions, stop/continue decisions, and answer tokens, using four trajectories per prompt and a shared trajectory-level group advantage. MSL-GRPO adds the early-sufficiency bonus and per-latent length cost to this policy.
The reward objectives are C: R = C; C+L: R = C − 0.1O/8; C+S: R = C(1 + 0.3S); Full: R = C(1 + 0.3S) − 0.1O/8. C is final-answer correctness; O is generated latent count. S = max(1 − k⋆/8, 0), with k⋆ the first prefix reaching correct-option probability 0.7 and k⋆ = 9 if there is no crossing. Thus, C+L removes sufficiency, C+S removes length cost, and C retains correctness alone.
The update uses standard GRPO group-relative advantages: each trajectory's reward is centered and scaled within its four-rollout group. Its latent actions, sampled stop decisions, and answer tokens all share that advantage. Equal rewards give zero policy advantage. At inference, the learned policy selects stop/continue directly from its logits.
SFT and every reward variant across all benchmarks. The SFT column is from Table 2; the reward columns are from Table 3. Scores are accuracy (%) ↑.
Table 3 summarizes the reward variants:
The fused-SFT native-generation analysis records 2.000 latent states per question over 13,266 questions. Full improves over fused SFT on all eight benchmark scores, including HR-4K 73.9→78.8, MME-RW 55.4→58.4, and VisualPuzzles 33.3→37.2. Against correctness-only C, Full gains 1.800 macro points while halving latent length. C+L reaches 72.066% with 1.700 latents and C+S 71.171% with 1.550; Full retains competitive macro accuracy with the shortest mean trajectory.
Does learned stopping help at inference? We keep the final RL checkpoint, answer decoder, and 700 questions fixed and compare learned stopping with exactly 2, 4, or 8 latent actions. The panel has 100 questions each from MMVP, HR-4K, HR-8K, MME-RW Lite, V*, VisualPuzzles, and MMStar. Table 5 reports:
Learned stopping answers 12–14 more questions correctly and reduces mean latency by 10.8% compared with eight fixed steps. The model switches to answer generation after its selected stop decision; fixed-length conditions switch after the prescribed number of latent transitions.
Manuscript changes. Table 2, p. 8 identifies C and the SFT baseline; §4.3, Tables 3 and 5, pp. 9–10 adds reward/length and stopping comparisons. §3.4, pp. 5–7 and Appendix E, Table 13, p. 33 specify the action blocks, advantage, and rewards. The revised manuscript is provided in the Supplementary Material with changes highlighted in blue.