{"paper":{"title":"ReSS: Learning Reasoning Models for Tabular Data Prediction via Symbolic Scaffold","license":"http://creativecommons.org/licenses/by/4.0/","headline":"ReSS extracts decision paths from trees to scaffold LLM fine-tuning for tabular prediction with faithful reasoning.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chenlang Yi, Gang Li, My T. Thai, Tianbao Yang, Tue Minh Cao, Yanmin Gong, Zizhan Xiong","submitted_at":"2026-04-15T01:43:00Z","abstract_excerpt":"Tabular data remains prevalent in high-stakes domains such as healthcare and finance, where predictive models are expected to provide both high accuracy and faithful, human-understandable reasoning. While symbolic models offer verifiable logic, they lack semantic expressiveness. Meanwhile, general-purpose LLMs often require specialized fine-tuning to master domain-specific tabular reasoning. To address the dual challenges of scalable data curation and reasoning consistency, we propose ReSS, a systematic framework that bridges symbolic and neural reasoning models. ReSS leverages a decision-tree"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental results on medical and financial benchmarks demonstrate that ReSS-trained models improve traditional decision trees and standard fine-tuning approaches up to 10% while producing faithful and consistent reasoning.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That decision-tree paths extracted from the data can serve as sufficient and non-restrictive scaffolds that force an LLM to generate reasoning which is both logically faithful and semantically useful without introducing new inconsistencies or losing predictive power.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ReSS uses decision-tree scaffolds to fine-tune LLMs for faithful tabular reasoning, reporting up to 10% gains over baselines on medical and financial data.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"ReSS extracts decision paths from trees to scaffold LLM fine-tuning for tabular prediction with faithful reasoning.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4099be7debb31ad76b0ffb33047c33fe9d58ec6e13195c5a1bf442e2d5568245"},"source":{"id":"2604.13392","kind":"arxiv","version":2},"verdict":{"id":"74b06c85-eee2-4fad-bb81-318d69c20d0e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T13:54:04.699619Z","strongest_claim":"Experimental results on medical and financial benchmarks demonstrate that ReSS-trained models improve traditional decision trees and standard fine-tuning approaches up to 10% while producing faithful and consistent reasoning.","one_line_summary":"ReSS uses decision-tree scaffolds to fine-tune LLMs for faithful tabular reasoning, reporting up to 10% gains over baselines on medical and financial data.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That decision-tree paths extracted from the data can serve as sufficient and non-restrictive scaffolds that force an LLM to generate reasoning which is both logically faithful and semantically useful without introducing new inconsistencies or losing predictive power.","pith_extraction_headline":"ReSS extracts decision paths from trees to scaffold LLM fine-tuning for tabular prediction with faithful reasoning."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.13392/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}