{"paper":{"title":"Teaching and Evaluating LLMs to Reason About Polymer Design Related Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Small language models trained on PolyBench match or beat larger models on polymer design reasoning tasks.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Benjamin Hsiao, Dikshya Mohanty, Mohammad Saqib Hasan, Niranjan Balasubramanian, Size Zheng, Syed Mostofa Monsur","submitted_at":"2026-01-22T20:39:18Z","abstract_excerpt":"Research in AI4Science has shown promise in many science applications, including polymer design. However, current LLMs are ineffective in this problem space because: (i) most models lack polymer-specific knowledge, and (ii) existing aligned models have limited coverage of knowledge and capabilities relevant to polymer design. Addressing this, we introduce PolyBench, a large-scale training and test benchmark dataset of more than 125K polymer design-related tasks, leveraging a knowledge base of more than 13 million data points obtained from experimental and synthetic data sources to ensure broad"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"small language models (SLMs) with 7B to 14B parameters, trained on PolyBench, outperform similar-sized models and remain competitive with closed-source frontier LLMs on PolyBench's test dataset, while demonstrating performance gains on external polymer benchmarks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The tasks in PolyBench and the underlying knowledge base of 13 million data points accurately represent the knowledge and reasoning demands of real-world polymer design without major gaps, biases, or inaccuracies from the experimental and synthetic sources.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Small 7B-14B parameter language models trained on the new PolyBench dataset for polymer design tasks outperform similar-sized models and compete with large frontier LLMs while improving on external benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Small language models trained on PolyBench match or beat larger models on polymer design reasoning tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e0017d5f117bfced1a30a9a94f4b66f10767b0397058ed71bedc6e05c5da0f27"},"source":{"id":"2601.16312","kind":"arxiv","version":2},"verdict":{"id":"6a5c5619-10db-4c94-9197-016d580cb279","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T11:31:08.017151Z","strongest_claim":"small language models (SLMs) with 7B to 14B parameters, trained on PolyBench, outperform similar-sized models and remain competitive with closed-source frontier LLMs on PolyBench's test dataset, while demonstrating performance gains on external polymer benchmarks.","one_line_summary":"Small 7B-14B parameter language models trained on the new PolyBench dataset for polymer design tasks outperform similar-sized models and compete with large frontier LLMs while improving on external benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The tasks in PolyBench and the underlying knowledge base of 13 million data points accurately represent the knowledge and reasoning demands of real-world polymer design without major gaps, biases, or inaccuracies from the experimental and synthetic sources.","pith_extraction_headline":"Small language models trained on PolyBench match or beat larger models on polymer design reasoning tasks."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"bafe42996f3a0f8bcc9c7d6a962d13e8a8bc0170e8a6d33427bf00128be8f92b"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}