{"id":"b8912ad8-737d-430e-a85e-3f8c06357e6a","arxiv_id":"2607.26018","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"A simple 'deletions' method — cubic extraction plus removal of atoms colliding across periodic boundaries — outperformed five alternative extraction methods, including a generative diffusion approach, for reproducing DFT forces in a-SiO2, dislocated Ta, and molten C.","lead":"Benchmarking six methods for cutting small periodic cells out of large atomistic simulations for DFT relabeling in active learning, this paper finds the simplest approach — extract a cube, apply periodic boundaries, delete atoms that collide — reproduces quantum forces best. The result simplifies how machine-learned interatomic potentials can be trained on structures that only exist at scale.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Deletions is not shown to be uniquely optimal: its force advantage over cubic extract is not statistically significant, and the energy-based tie-breaker is admitted to be non-rigorous and also penalizes deletions for Ta.","rationale":"The paper is a careful benchmark with transparent caveats: multiple materials, DFT reference forces, paired t-tests, and an honest admission that energy comparisons are not rigorous. The central proposal—that simple deletion of boundary-colliding atoms can yield DFT-tenable environments—is plausible and moderately supported. However, the headline assertion of 'superior performance' and 'optimal' overstates the evidence. The force metric, the only rigorous one, does not distinguish deletions from cubic extract, and the energy tie-breaker is both non-rigorous and internally inconsistent for Ta. The F_tol tuning via source DFT forces is an additional practical limitation, though the paper scopes its claim to the tested systems. The reader's CONDITIONAL verdict is appropriate; my stress-test identifies the same soft spot but frames it as an evidentiary gap rather than solely a threshold-transferability issue, hence 'partial' agreement. No verdict change is needed.","tokens_in":12493,"tokens_out":6271,"duration_ms":111448,"concrete_test":"Perform a paired non-inferiority test on force RMSE between deletions and cubic extract for all three materials, with a pre-specified margin (e.g., 0.01 eV/Å), and simultaneously compute a rigorous energy comparison: for each extracted environment, compare DFT energies of deletions and cubic-extract configurations after removing boundary close contacts by a simple distance cut (no IAP force threshold), using the source configuration's per-atom energy and stoichiometry as reference. If the force RMSEs are within the non-inferiority margin and the energy penalty of cubic extract after distance-based cleanup is comparable to deletions, then the central 'optimal' claim should be downgraded to 'among the best.'","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—deletions is the optimal extraction method among those considered—is not established by the paper's own evidence. On the primary force metric (Sec. III.A), deletions and cubic extract have statistically indistinguishable RMSEs for SiO2 (p=0.80) and 'equivalent RMSE' for C; for Ta, deletions is not significantly better than generative (p=0.08) or deletions+relax (p=0.16). The only basis for preferring deletions over cubic extract is the per-atom energy comparison in Sec. III.D/Fig. 8. But the authors explicitly state that comparing DFT total energies of extracted cells to the source is 'not rigorous' (Sec. III.D), and the same energy metric shows deletions leaves residual close contacts in Ta with energies ~8 eV/atom above source (Sec. III.C, Fig. 8b). Thus, by the paper's own criterion, deletions is also energetically contaminated, undermining the energy-based dismissal of cubic extract. Additionally, the deletion threshold F_tol is chosen per material as 'close to, but greater than, the maximum DFT force in the respective source configuration' (Table I), meaning the method relies on knowledge of the reference DFT forces it is supposed to avoid computing. The data support deletions as one of the best force-preserving methods, but not as uniquely optimal.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper benchmarks six methods for extracting atom-centered environments from large bulk configurations into smaller periodic cells suitable for DFT, with the goal of reproducing the DFT forces of the extracted atoms. The methods include the previously reported anneal (AME) approach and several new ones: spherical extract, cubic extract, deletions, deletions+relax, and a generative diffusion-model-based approach. Three materials are tested: amorphous SiO2, BCC Ta with screw dislocations, and molten C. The evaluation uses independent DFT reference forces on atoms in the source configurations and compares the forces in the extracted cells against those references. The authors report that the simple deletions method—which removes high-force atoms across periodic boundaries after a cubic extract—yields the lowest or tied-lowest force RMSE and maximum error for the central atoms among all methods, and they assert that deletions is the optimal extraction method because cubic extract, while competitive on forces, produces unphysically large per-atom energies.","tokens_in":12842,"tokens_out":2590,"duration_ms":42120,"significance":"If the central claim were fully established, the paper would provide a valuable practical recommendation: a trivial boundary-cleanup rule after cubic extraction can match or exceed more sophisticated generative or annealing-based methods for preparing MLIP training data, at a fraction of the computational cost. The study is carefully executed on three materially diverse systems, with explicit DFT convergence testing and paired t-tests on the force errors, which is a methodological improvement over prior anecdotal comparisons. The deletions method is also simple enough to be immediately adopted by practitioners. However, the force-accuracy advantage of deletions over cubic extract is not statistically significant for SiO2 and is only marginal for Ta relative to generative and deletions+relax, so the paper's headline claim of optimality rests heavily on a per-atom energy comparison that the authors themselves acknowledge is not rigorous and that also shows deletions leaves residual close contacts in Ta. The practical insight is still useful, but the strength of the conclusion exceeds what the evidence supports.","major_comments":[{"comment":"The central claim—that deletions is the optimal extraction method—is not supported by the paper's own primary metric. The paired t-tests in §III.A give p=0.80 for the SiO2 force-error difference between deletions and cubic extract, and p=0.08 and p=0.16 for Ta versus generative and deletions+relax, respectively, none of which reach the p≤0.05 significance criterion. For C, the RMSEs are reported as equivalent. Thus the force data show deletions is tied with cubic extract, not superior. The conclusion in §IV overstates the evidence; the claim should be softened to 'among the best and simplest' unless additional evidence is provided.","section":"§IV and §III.A"},{"comment":"The energy-based tie-breaker used to dismiss cubic extract is not internally consistent. The authors state in §III.D that comparing DFT total energies of extracted cells to the source is 'not rigorous,' yet they use the per-atom energy excess of cubic extract (especially for Ta) as the decisive reason to prefer deletions. The same metric shows deletions itself leaves residual close contacts in Ta, with energies ~8 eV/atom above the source (Fig. 8b, inset). By the paper's own criterion, deletions is also energetically contaminated for Ta. If the energy metric is allowed to rule out cubic extract, it equally undermines deletions for Ta. The conclusion should either rely on force-only metrics (which do not distinguish deletions from cubic extract) or justify a rigorous energy-decomposition framework.","section":"§III.D and Fig. 8"},{"comment":"The deletions method's performance depends on the per-material force threshold F_tol, which is chosen as 'close to, but greater than, the maximum DFT force in the respective source configurations' (Table I). This means the method requires prior knowledge of the DFT force distribution of the large source configuration—exactly the quantity the extraction procedure is meant to avoid computing. No sensitivity analysis is provided to show that the results are robust to reasonable variations in F_tol, nor is an a priori selection rule given. Since F_tol is effectively tuned to each source, the claim that deletions is broadly superior may not transfer to new systems where the source DFT forces are unknown. A sensitivity study or a transferable prescription for setting F_tol is needed to support the general recommendation.","section":"Table I and §II.A (deletions)"}],"minor_comments":[{"comment":"The paired t-test is described only in the main text; the appendix shows figures but not the numerical p-values for all pairs. Including a table of p-values for all method pairs and materials would make the statistical analysis more transparent and reproducible.","section":"Appendix C"},{"comment":"Several table captions in Appendix A are mislabeled: Tables A3 and A4 both describe Ta, but A4 is captioned 'k-points mesh size' while its column is h [Å], and A5/A6 similarly mix k-mesh and real-space mesh descriptions. Correct the captions to avoid confusion.","section":"Table captions (A3–A6)"},{"comment":"The abstract says 'We demonstrated a notably simple procedure'—change to 'We demonstrate' for consistency with the rest of the paper's tense.","section":"Abstract"},{"comment":"The anneal method description mentions a margin δ_margin=0.5 Å and removal of atoms to achieve stoichiometry for SiO2, but the subsequent density analysis in the SI shows the anneal configurations have densities 33% lower than the source. A brief comment on how the margin and stoichiometry adjustments affect density would help the reader interpret the structural results.","section":"§II.A (anneal)"},{"comment":"The RDF plot for cubic extract is said to contain an extraneous peak near r=0, but this is difficult to see in the main figure. The zoomed-in panels are in the SI; consider adding a zoomed inset in the main figure or citing the SI more prominently at first mention.","section":"Fig. 5"}],"recommendation":"major_revision","confidential_remarks":"The paper is a solid comparative benchmark with careful DFT convergence testing and statistical analysis, but the headline claim of optimality for the deletions method is not supported by the force-based evidence alone. The energy-based tie-breaker is both non-rigorous and also penalizes deletions for Ta, creating an inconsistency in the argument. The work can likely be revised to a publishable state by softening the conclusion and adding the requested sensitivity analysis for F_tol, or by providing a more rigorous energy metric. I do not see a fatal flaw in the experimental design, so rejection is not warranted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper does something genuinely useful: it benchmarks six ways to extract local atomic environments from large MD cells for DFT training data, and finds that a trivial procedure—cut out a cube, then delete atoms that collide across the periodic boundaries—matches or beats fancier generative and annealing methods on DFT force reproduction. That is a practical result worth knowing, and the evaluation is more careful than most work in this area: direct DFT force comparisons, convergence testing, paired t-tests, and a sensible spread of materials (amorphous SiO2, dislocated Ta, molten C). The authors also make a fair point that force errors, not total energies, are the right primary metric.\n\nWhere the paper oversells itself is the conclusion that deletions is 'the optimal extraction method.' On the main force metric, deletions' advantage over cubic extract is not statistically significant—p=0.80 for SiO2, and the authors admit 'equivalent RMSE' for C. For Ta, the advantage over generative and deletions+relax is marginal. The tie-breaker used to prefer deletions over cubic extract is the per-atom energy comparison, which the authors themselves call 'not rigorous.' And on that same energy metric, deletions leaves residual close contacts in Ta with ~8 eV/atom excess energy—better than cubic extract's hundreds of eV/atom, but still a clear sign that the deletion procedure isn't clean. So the honest statement is: deletions matches cubic extract on forces while removing the most egregious boundary collisions, at essentially zero computational cost. That is worth saying, but it is not the same as 'demonstrated superior performance.'\n\nA few smaller soft spots: F_tol is chosen per material by looking at the maximum DFT force in the source, which means the method's one parameter is fit to the very calculation it is meant to avoid; no code or data are released; and the GRS method, praised in the introduction, is never benchmarked against the others. These are all addressable in revision.\n\nNet: this is a solid, reproducible contribution to MLIP data curation, and the deletions method should be in active-learning toolkits. But the conclusion needs rebalancing. A serious referee would push on the statistics and the F_tol dependence, and the paper would come out stronger.\n\nRecommendation: send to peer review. It deserves a careful referee, and with moderate revisions it will be a useful reference.","headline":"Useful benchmark and a pleasantly simple winner, but the 'optimal' claim outruns the statistics.","tokens_in":13337,"tokens_out":1443,"would_cite":true,"duration_ms":24753,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A cubic cut followed by deleting colliding atoms reproduces DFT forces as well as or better than generative and annealing methods, making it the preferred way to prepare MLIP training data.","keywords":["atomic environment extraction","machine learning interatomic potentials","periodic boundary conditions","DFT training data","active learning","diffusion model","deletions method","amorphous SiO2"],"falsifier":"Run the deletions protocol on a material with intrinsically large internal forces (e.g., a covalently bonded glass) using the paper's F_tol rule, and compare DFT forces of the central atom against the source; if the RMSE exceeds that of cubic extract by more than the paper's measured gaps, the claim of universality fails. A cheaper check: scan F_tol over a range for SiO2 and show whether the optimal threshold lands at the paper's chosen value or at a value that depends on the arbitrary source frame.","tokens_in":12377,"feed_emoji":"⚛️","tokens_out":4181,"duration_ms":59557,"temperature":0.7,"pith_summary":"This paper tackles a practical bottleneck in building machine-learned interatomic potentials: carving a small, DFT-tractable cell out of a large simulation box without corrupting the forces of the atoms one cares about. The authors benchmark six extraction strategies across three materials — amorphous silica, dislocated tantalum, and molten carbon — and find that the most low-tech option wins. The 'deletions' method simply cuts a cubic cell around an atom of interest, applies periodic boundary conditions, and sequentially removes atoms whose MLIP-evaluated forces exceed a threshold, eliminating collisions across the new boundary. On force accuracy, deletions is best or tied for best in every material; its advantage over the runner-up cubic cut is not statistically significant, while its advantage over diffusion and annealing methods is. Because it needs no bespoke AI model and no long annealing runs, the paper argues deletions should be the default first choice for extraction, even though it leaves residual boundary defects and slightly inflated energies in tantalum.","feed_headline":"Deleting boundary atoms beats AI methods for MLIP data extraction","feed_subtitle":"A cubic cut plus force-based atom deletion matches or beats diffusion and annealing across SiO2, Ta, and molten C.","key_machinery":"The deletions algorithm: (1) choose a fixed spherical core of radius r_core around a central atom; (2) cut out the entire cube of side L_cell from the source; (3) apply periodic boundary conditions; (4) using a pretrained IAP, compute per-atom forces, and repeatedly remove the outside-core atom with the largest |F| until max |F| < F_tol. The workhorse insight is that boundary collisions manifest as outlier forces, so force magnitude is a sufficient collision detector; no relaxation, no generation, and no stoichiometry fixing is needed.","core_discovery":"The central claim is that, for bulk systems, the atomic forces of a preserved environment are best reproduced when the extraction method keeps the central region's local geometry exactly as it was in the source, even at the cost of structural inconsistency at the newly imposed periodic boundaries. The deletions method achieves this: it extracts all atoms in a cube around the central atom, applies PBCs, and iteratively deletes the atom with the largest force outside the fixed core, as judged by a cheap interatomic potential, until the maximum force falls below a per-material tolerance. Across SiO2, Ta, and C, deletions matched or beat every alternative — including a diffusion-model-based gene","pith_inferences":["Since the superiority of deletions over cubic extract is not statistically distinguishable on force error alone, the decisive advantage is energy cleanliness; the conclusion would flip if a way to neutralize cubic-extract collisions (e.g., short relaxation) produced identical forces.","The per-material F_tol tuning rule (just above max source force) is a heuristic; testing a fixed universal threshold across materials would clarify whether the method is effectively parameter-free in practice.","If force-magnitude thresholds are transferable, the same recipe could be applied to other embeddings (spherical clusters in vacuum, slab cells) without re-tuning, but that remains untested.","The finding suggests that 'physical plausibility' of the boundary region (as produced by diffusion or annealing) matters less than exact preservation of near-neighbor geometry; this invites a test of how much boundary disorder a trained MLIP actually tolerates."],"forward_implications":["For bulk MLIP training-set construction, the deletions method gives DFT-force accuracy on par with or better than generative and annealing approaches, at a fraction of the setup cost.","Because deletions requires only an approximate force evaluator, it could be run with a simple distance cutoff instead of an MLIP; the paper explicitly suggests this.","The per-atom energy penalty of cubic extract (hundreds of eV in Ta) is removed by deletions, making the latter usable for energy-sensitive training.","Force accuracy degrades with distance from the central atom in all methods; deletions and cubic extract degrade most gradually, so a single DFT cell can provide trustworthy forces for more atoms.","The method is directly useful for active-learning and on-the-fly training workflows where large-scale MD configurations must be repeatedly reduced to DFT-sized cells."],"fun_headline_variants":["Deletion method beats AI for extracting MLIP training data","Keep core geometry, delete outlier atoms: best DFT extraction","Simple deletions outmatch diffusion and annealing for atomic samples","Force-based atom deletion wins for MLIP data extraction","Cubic-cut deletions outperform AI for atomic environment harvesting"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The whole recommendation rests on the assumption that a single per-material force threshold F_tol, set just above the maximum DFT force in the source configuration, reliably identifies the atoms whose removal fixes boundary collisions without disturbing the preserved core — an assumption the Ta results already strain, since residual collisions remain and the per-atom energy is still inflated.","fun_headline_variants_meta":{"raw":{"variants":["Deletion method beats AI for extracting MLIP training data","Keep core geometry, delete outlier atoms: best DFT extraction","Simple deletions outmatch diffusion and annealing for atomic samples","Force-based atom deletion wins for MLIP data extraction","Cubic-cut deletions outperform AI for atomic environment harvesting"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000347,"raw_usage":{"total_tokens":1745,"prompt_tokens":761,"completion_tokens":984,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":505,"completion_tokens_details":{"reasoning_tokens":906}},"tokens_in":505,"tokens_out":984,"duration_ms":13866,"temperature":1.0,"reasoning_tokens":906,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T03:12:33.277625+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the deletions protocol on a material with intrinsically large internal forces (e.g., a covalently bonded glass) using the paper's F_tol rule, and compare DFT forces of the central atom against the source; if the RMSE exceeds that of cubic extract by more than the paper's measured gaps, the claim of universality fails. A cheaper check: scan F_tol over a range for SiO2 and show whether the optimal threshold lands at the paper's chosen value or at a value that depends on the arbitrary source frame.","supporting_citations":[],"review_version":2}