- Rating
- 1
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
The paper presents PROGEA, a unified diffusion language model that integrates protein sequences, discretized backbone structures, and natural language through a shared token vocabulary. The authors compare MDLM and BDLM across cross-modal tasks, providing an empirical assessment of their respective strengths in structural modeling, sequence design, and protein-domain language interaction.
Critical strengths
-
The native integration of sequence, structure, and language within a shared dLLM backbone is a valuable systems contribution. The MDLM-BDLM comparison provides evidence of task-dependent trade-offs between fully noncausal and blockwise generation.
-
PROGEA-MDLM achieves substantially better folding metrics than the reevaluated trimodal AR baseline, while PROGEA-BDLM attains GeneralQA accuracy numerically comparable to the evaluated strong general-purpose LLMs.
Critical weaknesses
-
Multi-task coverage does not establish positive transfer. The experiments lack controlled single-task, bimodal, and trimodal comparisons and task or modality removal ablations. Without matching the backbone, target-task data exposure, and training budget, the marginal benefit of joint training remains unidentified. The reported gains therefore cannot be disentangled from additional supervision or regularization effects.
-
The contribution of diffusion-based generation remains confounded. The MDLM-BDLM comparison and attention analysis support the proposed mechanism, but do not establish that diffusion-based generation itself explains the improvements over the AR baseline.
-
The incremental methodological contribution of fusion is unclear. Token-level early fusion and staged modality alignment are described, but their novelty and independent performance contributions relative to prior shared-vocabulary models remain unresolved. Figure 4 provides descriptive evidence about representation geometry, rather than causal evidence of task-level gains.
-
The incremental utility of a unified trimodal system remains insufficiently validated. Evaluation primarily covers single-step mappings between modality pairs. Tasks jointly constrained by sequence, structure, and text are underexplored, and multi-stage generation lacks systematic evaluation of cross-modal consistency, error accumulation, compositional generalization, and end-to-end success.
-
The framework figure (Figure 2) has limitations in both scientific accuracy and methodological clarity. Several regions in the protein illustrations appear structurally implausible, and the folding landscape risks conflating conditional backbone prediction with physical folding dynamics. The figure also insufficiently highlights the framework’s distinctive features, particularly the attention visibility patterns and denoising procedures that distinguish MDLM from BDLM. More accurate structural illustrations and explicit depictions of these mechanisms would better communicate the framework’s technical contributions.
Questions for authors
-
Does joint training yield positive transfer beyond multi-task coverage? Please provide controlled single-task, bimodal, and trimodal comparisons with matched backbone, target-task supervision, and training compute, together with task or modality removal ablations.
-
Which components constitute the incremental contribution over prior shared-vocabulary trimodal models, and what controlled ablations isolate their effects on downstream performance?
-
Could the authors extend the structure prediction evaluation to CASP14 and CASP15 alongside the reported CAMEO22 results? Comparisons with established structure prediction methods on these benchmarks would provide a more comprehensive assessment of the model’s structural prediction capability.
-
Given that PROGEA’s sequence-text supervision and GeneralQA both derive from Swiss-Prot, could the authors clarify the train-test decontamination procedures and provide the complete evaluation prompts, specifying any information supplied beyond the sequence and question?
-
For the Claude and GPT baselines listed in Table 3, could the authors clarify the prompting strategy, reasoning and output token budgets, decoding settings, and whether tool use or external retrieval was enabled?