{"paper":{"title":"Bridging the Last Mile of Circuit Design: PostEDA-Bench, a Hierarchical Benchmark for PPA Convergence and DRC Fixing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"LLM agents succeed on simple post-EDA tasks but reach only 37% and 20% success on practical DRC reasoning and multi-objective PPA.","cross_cats":["cs.AI","cs.MA"],"primary_cat":"cs.AR","authors_text":"Caiwen Ding, Jinwei Tang, Nuo Xu, Pengju Liu, Yu Cao","submitted_at":"2026-05-07T20:54:07Z","abstract_excerpt":"LLM-based agents are increasingly applied to the \"last mile\" of Electronic Design Automation (EDA): repairing residual sign-off Design Rule Check (DRC) violations and converging Power-Performance-Area (PPA) targets after tool runs. Existing EDA-LLM benchmarks, however, omit DRC fixing entirely and rely on flat hierarchies tied to a single toolchain. We introduce PostEDA-Bench, a hierarchical benchmark with 145 tasks across DRC-Essential, DRC-Reasoning, PPA-Mono, and PPA-Multi, supported by EDA toolchains with machine-checkable evaluation. Across eight commercial and open-source LLMs under mult"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across eight commercial and open-source LLMs under multiple agent scaffolds, we find that agents handle synthetic DRC-Essential and single-objective PPA-Mono reasonably well but degrade sharply on the more practical DRC-Reasoning, where the best success rate is 36.66%, and PPA-Multi, where the best success rate is 20.00%; vision augmentation consistently enhances DRC-Bench; and trade-off reasoning, rather than knob knowledge, is the dominant PPA-Multi bottleneck.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The 145 tasks across the four categories in PostEDA-Bench, supported by EDA toolchains with machine-checkable evaluation, are representative of real-world post-EDA challenges and provide a fair test of agent capabilities.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"PostEDA-Bench shows LLM agents succeed reasonably on basic DRC and single-objective PPA tasks but struggle on practical DRC reasoning (best 36.66% success) and multi-objective PPA (best 20% success).","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM agents succeed on simple post-EDA tasks but reach only 37% and 20% success on practical DRC reasoning and multi-objective PPA.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"031218625567234a91d6c30074af8ca56e7be852b08704fd8a9149115488ea10"},"source":{"id":"2605.06936","kind":"arxiv","version":2},"verdict":{"id":"d9a1ffce-a062-430b-9190-50a9ca2f92ea","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-11T00:57:42.284462Z","strongest_claim":"Across eight commercial and open-source LLMs under multiple agent scaffolds, we find that agents handle synthetic DRC-Essential and single-objective PPA-Mono reasonably well but degrade sharply on the more practical DRC-Reasoning, where the best success rate is 36.66%, and PPA-Multi, where the best success rate is 20.00%; vision augmentation consistently enhances DRC-Bench; and trade-off reasoning, rather than knob knowledge, is the dominant PPA-Multi bottleneck.","one_line_summary":"PostEDA-Bench shows LLM agents succeed reasonably on basic DRC and single-objective PPA tasks but struggle on practical DRC reasoning (best 36.66% success) and multi-objective PPA (best 20% success).","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The 145 tasks across the four categories in PostEDA-Bench, supported by EDA toolchains with machine-checkable evaluation, are representative of real-world post-EDA challenges and provide a fair test of agent capabilities.","pith_extraction_headline":"LLM agents succeed on simple post-EDA tasks but reach only 37% and 20% success on practical DRC reasoning and multi-objective PPA."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.06936/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T11:42:03.512658Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T06:39:35.896396Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T17:31:19.310570Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:14:10.140053Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"c804961ceb7f651adb76397e7b89962bff64e1ac7ab6880cc33bb4b4c1e87c46"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"eb2cd9baa6597817a8e41f2eba9af508e09d57c54ee63a6d494b35294db17a8b"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}