{"id":"9eb8da93-3c53-4572-9b37-893d5b5243a8","arxiv_id":"2604.24310","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Vib2Conf achieves over 95% top-1 recall on standard spectrum-to-structure benchmarks and 82% recall for distinguishing near-isomeric 3D conformers differing by only ~1 Å RMSD.","lead":"Vib2Conf is a deep learning model that maps vibrational spectra directly to 3D molecular conformations using an attentional resampler and Mixture-of-Experts architecture. A smart generalist might read it to understand how AI can resolve fine structural details from spectroscopic data that were previously ambiguous.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Performance on near-isomeric conformers (RMSD ~1Å) may not hold under experimental spectral noise and solvent effects absent from training data.","rationale":"The reader's weakest assumption matches the identified load-bearing point exactly. The paper demonstrates strong benchmark results on synthetic data, but the leap to 'fine-grained spectrum-to-conformation analysis' in practice rests on untested generalization. This moves the verdict from UNVERDICTED to CONDITIONAL pending experimental validation; the internal benchmark claims remain intact.","tokens_in":1724,"tokens_out":372,"duration_ms":45748,"concrete_test":"Apply the trained Vib2Conf model (or a retrained version) to a held-out set of 15–20 experimental IR/Raman spectra for molecules with crystallographically or spectroscopically confirmed multiple conformers (e.g., selected from NIST or published studies on alkanes and small organics); compute top-1 recall on the experimental set and compare to the 82% synthetic figure. A drop >20% indicates the concern lands.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim requires that conformation-sensitive features extracted by the attentional resampler and MoE are robust enough to distinguish subtle spectral differences between near-isomers. This hinges on the training distributions in QM9S, VB-Mols, QMe14S, and VB-Confs capturing real conformational heterogeneity and noise. These sets are generated from quantum-chemical calculations (typically DFT in vacuum), yielding idealized, noise-free spectra. Real experimental vibrational spectra introduce peak broadening, baseline drift, solvent-induced shifts, and anharmonicities that can mask the small differences at ~1Å RMSD. If the model exploits dataset-specific artifacts rather than physically invariant features, the 82.06% top-1 recall will not generalize.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript introduces Vib2Conf, a deep learning model that uses an attentional resampler to extract conformation-sensitive features from vibrational spectra and a Mixture-of-Experts (MoE) module to map these to 3D conformations. It reports state-of-the-art top-1 recall exceeding 95% on the QM9S, VB-Mols, and QMe14S benchmarks, and a top-1 recall of 82.06% on the VB-Confs test set for distinguishing near-isomeric conformers that differ by an RMSD of only ~1 Å.","tokens_in":1882,"tokens_out":606,"duration_ms":65389,"significance":"If the performance claims hold under scrutiny, the work would represent a meaningful advance in spectrum-to-structure retrieval by addressing conformational ambiguity, which remains a core limitation in vibrational spectroscopy applications. The attentional resampler and MoE integration provide a plausible mechanism for handling sparse signals and partitioning conformational space. The reported results on multiple benchmarks, including the challenging near-isomer case, constitute a strength, though the idealized DFT-derived training data limit immediate claims of experimental utility.","major_comments":[{"comment":"Results section on VB-Confs benchmark: The central claim of 82.06% top-1 recall for ~1 Å RMSD conformers is load-bearing, yet no ablation studies are presented that isolate the contribution of the attentional resampler versus the MoE (or a baseline without either); without these, it is impossible to confirm that the architecture, rather than dataset artifacts, drives the discrimination of subtle spectral differences.","section":"Results (VB-Confs)"},{"comment":"Methods section describing datasets and training: No details are provided on train/test splits, cross-validation strategy, random seeds, or any spectral noise augmentation for QM9S, VB-Mols, QMe14S, or VB-Confs; this omission directly affects assessment of whether the 82.06% recall reflects generalization or potential overfitting to noise-free DFT spectra.","section":"Methods"}],"minor_comments":[{"comment":"Abstract: The statement 'exceeding 95%' on the three benchmarks would be more informative if the exact per-dataset recalls were stated rather than aggregated.","section":"Abstract"},{"comment":"Notation: The RMSD threshold of '~1 Å' is used without a precise definition or distribution statistics for the VB-Confs pairs; adding a table or histogram of RMSD values would clarify the difficulty of the test cases.","section":"Results"}],"recommendation":"major_revision","confidential_remarks":"The manuscript would be strengthened by releasing code, trained models, and exact data splits to enable independent verification, especially given the AI-driven nature of the claims. The citation list appears balanced, but the work sits at the intersection of chemoinformatics and spectroscopy; confirm fit with journal scope."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their constructive and detailed review of our manuscript. We address each major comment point by point below. Where the comments identify omissions that affect the strength of our claims, we commit to revisions that will incorporate the requested information and analyses.","responses":[{"response":"We agree that ablation studies are necessary to rigorously attribute the 82.06% top-1 recall on VB-Confs to the attentional resampler and MoE components rather than dataset properties. In the revised manuscript we will add a dedicated ablation subsection in the Results, reporting performance of the full model against (i) a variant without the attentional resampler, (ii) a variant without the MoE module, and (iii) a simple baseline without either component. These experiments will be performed on the same VB-Confs test set and will be accompanied by statistical significance tests.","revision_made":"yes","referee_comment":"[Results (VB-Confs)] Results section on VB-Confs benchmark: The central claim of 82.06% top-1 recall for ~1 Å RMSD conformers is load-bearing, yet no ablation studies are presented that isolate the contribution of the attentional resampler versus the MoE (or a baseline without either); without these, it is impossible to confirm that the architecture, rather than dataset artifacts, drives the discrimination of subtle spectral differences."},{"response":"We thank the referee for highlighting this gap. The revised Methods section will explicitly state the train/test split ratios and selection criteria for each benchmark, the cross-validation strategy employed, the random seeds used for all experiments (to ensure reproducibility), and whether spectral noise augmentation was applied (none was used, as all spectra were generated from noise-free DFT calculations). These additions will allow readers to evaluate generalization versus potential overfitting.","revision_made":"yes","referee_comment":"[Methods] Methods section describing datasets and training: No details are provided on train/test splits, cross-validation strategy, random seeds, or any spectral noise augmentation for QM9S, VB-Mols, QMe14S, or VB-Confs; this omission directly affects assessment of whether the 82.06% recall reflects generalization or potential overfitting to noise-free DFT spectra."}],"tokens_in":1424,"tokens_out":481,"duration_ms":57116,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"Vib2Conf reaches 82% top-1 recall on near-isomeric conformers at ~1 Å RMSD using an attentional resampler and MoE, but all results sit on clean simulated spectra. The architecture appears tailored to pull conformation-sensitive signals from sparse vibrational data and to split the space across experts, which lets it beat prior numbers on the standard QM9S, VB-Mols, and QMe14S sets with over 95% top-1 recall. That part is concrete and worth looking at if you work on spectrum-to-structure mapping. The paper also ships a new test set, VB-Confs, focused on close isomers, so the evaluation is not just re-running old benchmarks. The central numbers come from held-out splits rather than circular fitting, which is a plus. The soft spot is the data source. Everything is generated from DFT in vacuum, so the spectra are noise-free and lack solvent shifts, peak broadening, or anharmonic effects that appear in real measurements. If the model is learning dataset artifacts instead of invariant features, the 82% figure will drop once experimental spectra are used. No ablation tables or noise-injection tests are described in the abstract, and the full text would need to show whether those checks were done. This paper is for computational chemists who already use deep learning for IR/Raman interpretation and want a practical next step on conformational resolution. A reader who needs a ready tool for drug-like molecules will find the numbers useful even if they later add their own experimental fine-tuning. It deserves a serious referee because the claim is specific, the architecture is described, and the evaluation uses distinct test sets; the generalization question is addressable in revision rather than fatal. Send it to review.","headline":"Vib2Conf reaches 82% top-1 recall on near-isomeric conformers at ~1 Å RMSD using an attentional resampler and MoE, but all results sit on clean simulated spectra.","tokens_in":2400,"tokens_out":434,"would_cite":false,"duration_ms":51426,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A deep learning model discriminates three-dimensional molecular conformations from vibrational spectra even for near-isomeric structures differing by roughly one angstrom in root-mean-square deviation.","keywords":["vibrational spectra","molecular conformations","deep learning","mixture of experts","attentional resampler","spectrum to conformation","conformational isomers","root mean square deviation"],"falsifier":"Measure vibrational spectra of a set of known near-isomeric conformers (RMSD ~1 Å) that were never seen during training and test whether Vib2Conf still returns the correct 3D structure as its top prediction.","tokens_in":2621,"feed_emoji":"🧪","tokens_out":716,"duration_ms":52357,"temperature":0.7,"pith_summary":"The paper introduces Vib2Conf to solve the problem of spectral ambiguities that arise when multiple three-dimensional conformations produce similar vibrational spectra. It builds an attentional resampler to pull out conformation-sensitive signals from sparse data and combines it with a Mixture-of-Experts layer that divides the space of possible shapes so each expert can map spectra to precise geometry. On standard benchmarks the model reaches top-1 recall above 95 percent, and on a dedicated set of near-isomers it still recovers the correct conformation 82 percent of the time. A reader would care because this moves spectrum-based structure retrieval from two-dimensional connectivity graphs toward usable three-dimensional atomic positions.","feed_headline":"Model maps vibrational spectra to specific 3D molecular shapes","feed_subtitle":"Vib2Conf reaches 82 percent top-1 recall on near-isomers that differ by only one angstrom RMSD, extending spectrum analysis to precise 3D ge","key_machinery":"Attentional resampler that extracts conformation-sensitive features from sparse spectral signals together with Mixture-of-Experts that partitions conformational space for geometric mapping.","core_discovery":"Vib2Conf directly discriminates 3D molecular conformations from vibrational spectra by distilling conformation-sensitive features from sparse signals with an attentional resampler and partitioning the conformational space with Mixture-of-Experts for precise geometric mapping, reaching state-of-the-art top-1 recall exceeding 95 percent on QM9S, VB-Mols, and QMe14S while achieving 82.06 percent top-1 recall on the VB-Confs test set for conformational isomers that differ by an RMSD of only approximately 1 Å.","pith_inferences":["The approach could reduce the need for complementary experimental techniques such as crystallography when only spectroscopic data are available.","Integration with existing computational workflows might allow real-time conformation assignment during spectroscopic experiments.","The method opens a route to test whether similar modular networks can resolve other ambiguous inverse problems in molecular spectroscopy."],"forward_implications":["Spectrum-to-structure retrieval extends from two-dimensional connectivity to direct three-dimensional conformational discrimination.","Conformers that differ by only about one angstrom RMSD become distinguishable from vibrational data alone.","The same attentional and expert-based architecture yields state-of-the-art recall on multiple established spectrum-structure benchmarks.","Fine-grained spectrum-to-conformation analysis becomes feasible for general use in molecular identification."],"fun_headline_variants":["Vib2Conf decodes 3D molecular conformations from vibrational spectra","AI separates near-isomeric 3D shapes from their vibrational spectra","Deep learning model resolves 3D conformations at 82 percent recall","Vib2Conf uses experts to map spectra to precise 3D geometries"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The training distributions from QM9S, VB-Mols, QMe14S, and VB-Confs sufficiently represent the conformational heterogeneity and spectral noise found in real experimental measurements.","fun_headline_variants_meta":{"raw":{"variants":["Vib2Conf decodes 3D molecular conformations from vibrational spectra","AI separates near-isomeric 3D shapes from their vibrational spectra","Deep learning model resolves 3D conformations at 82 percent recall","Vib2Conf uses experts to map spectra to precise 3D geometries"]},"model":"grok-4.3","cost_usd":0.010433,"raw_usage":{"total_tokens":4547,"prompt_tokens":694,"num_sources_used":0,"completion_tokens":76,"cost_in_usd_ticks":104328000,"prompt_tokens_details":{"text_tokens":694,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":3777,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":694,"tokens_out":76,"duration_ms":85641,"temperature":1.0,"reasoning_tokens":3777,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-07T17:29:15.832616+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Measure vibrational spectra of a set of known near-isomeric conformers (RMSD ~1 Å) that were never seen during training and test whether Vib2Conf still returns the correct 3D structure as its top prediction.","supporting_citations":[],"review_version":1}