{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BDDQ3SU5C4FNMMBT2LSZ5F54N4","short_pith_number":"pith:BDDQ3SU5","schema_version":"1.0","canonical_sha256":"08c70dca9d170ad63033d2e59e97bc6f28bb589f8e93b171468152d78bb4ad37","source":{"kind":"arxiv","id":"2605.12620","version":1},"attestation_state":"computed","paper":{"title":"Think Twice, Act Once: Verifier-Guided Action Selection For Embodied Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A test-time verifier trained on synthesized failures helps MLLM agents pick reliable actions from multiple candidates.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anna Rohrbach, Christian Bialas, Georgia Chalvatzaki, Marcus Rohrbach, Nishad Singhi, Snehal Jauhri, Vignesh Prasad","submitted_at":"2026-05-12T18:08:24Z","abstract_excerpt":"Building generalist embodied agents capable of solving complex real-world tasks remains a fundamental challenge in AI. Multimodal Large Language Models (MLLMs) have significantly advanced the reasoning capabilities of such agents through strong vision-language knowledge and chain-of-thought (CoT) reasoning, yet remain brittle when faced with challenging out-of-distribution scenarios. To address this, we propose Verifier-Guided Action Selection (VegAS), a test-time framework designed to improve the robustness of MLLM-based embodied agents through an explicit verification step. At inference time"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2605.12620","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T18:08:24Z","cross_cats_sorted":[],"title_canon_sha256":"df1ac8fdff8401667f177b8bd831d1159e04ca4b32cb55a1738c26c46f7d3af5","abstract_canon_sha256":"357ae961b7ad0cd33cf52aba688de6cfeb3ccedb2b2a1e58e3e3dc61b353b0ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:10:00.448101Z","signature_b64":"IN4Pv1TQibiSC+bPRIhs+huFCwXHepPs1IfjtYXeW+ZQf0QhGULhWWMa1SGkvGG4ofqR3XBVFpXBx0y/OghADQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"08c70dca9d170ad63033d2e59e97bc6f28bb589f8e93b171468152d78bb4ad37","last_reissued_at":"2026-05-18T03:10:00.447469Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:10:00.447469Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Think Twice, Act Once: Verifier-Guided Action Selection For Embodied Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A test-time verifier trained on synthesized failures helps MLLM agents pick reliable actions from multiple candidates.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anna Rohrbach, Christian Bialas, Georgia Chalvatzaki, Marcus Rohrbach, Nishad Singhi, Snehal Jauhri, Vignesh Prasad","submitted_at":"2026-05-12T18:08:24Z","abstract_excerpt":"Building generalist embodied agents capable of solving complex real-world tasks remains a fundamental challenge in AI. Multimodal Large Language Models (MLLMs) have significantly advanced the reasoning capabilities of such agents through strong vision-language knowledge and chain-of-thought (CoT) reasoning, yet remain brittle when faced with challenging out-of-distribution scenarios. To address this, we propose Verifier-Guided Action Selection (VegAS), a test-time framework designed to improve the robustness of MLLM-based embodied agents through an explicit verification step. At inference time"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Across embodied reasoning benchmarks spanning the Habitat and ALFRED environments, VeGAS consistently improves generalization, achieving up to a 36% relative performance gain over strong CoT baselines on the most challenging multi-object, long-horizon tasks.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That training a verifier on automatically synthesized failure cases from an LLM will produce a model that reliably identifies good actions in out-of-distribution scenarios where the base MLLM fails.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"VeGAS improves MLLM-based embodied agents by sampling action ensembles and using a verifier trained on LLM-synthesized failure cases, yielding up to 36% relative gains on hard multi-object long-horizon tasks in Habitat and ALFRED.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A test-time verifier trained on synthesized failures helps MLLM agents pick reliable actions from multiple candidates.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9e87c00acfd66d34ca16915d4712344a09642c67cd415fa10b35d95b88e72510"},"source":{"id":"2605.12620","kind":"arxiv","version":1},"verdict":{"id":"1e46b4c7-8745-451b-91f1-81cf25d3ab8e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:00:48.159839Z","strongest_claim":"Across embodied reasoning benchmarks spanning the Habitat and ALFRED environments, VeGAS consistently improves generalization, achieving up to a 36% relative performance gain over strong CoT baselines on the most challenging multi-object, long-horizon tasks.","one_line_summary":"VeGAS improves MLLM-based embodied agents by sampling action ensembles and using a verifier trained on LLM-synthesized failure cases, yielding up to 36% relative gains on hard multi-object long-horizon tasks in Habitat and ALFRED.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That training a verifier on automatically synthesized failure cases from an LLM will produce a model that reliably identifies good actions in out-of-distribution scenarios where the base MLLM fails.","pith_extraction_headline":"A test-time verifier trained on synthesized failures helps MLLM agents pick reliable actions from multiple candidates."},"references":{"count":65,"sample":[{"doi":"","year":2022,"title":"Do As I Can, Not As I Say: Grounding Language in Robotic Affordances","work_id":"037320f1-b0a9-4cbe-a639-bfb25409ce71","ref_index":1,"cited_arxiv_id":"2204.01691","is_internal_anchor":true},{"doi":"","year":2024,"title":"Critique-out-loud reward models","work_id":"adbb5d79-de43-49c4-abd5-ad004e77abea","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Qwen2.5-VL Technical Report","work_id":"69dffacb-bfe8-442d-be86-48624c60426f","ref_index":3,"cited_arxiv_id":"2502.13923","is_internal_anchor":true},{"doi":"","year":2024,"title":"Large Language Monkeys: Scaling Inference Compute with Repeated Sampling","work_id":"b124064d-5a56-42ad-86f5-3cc349b86a3a","ref_index":4,"cited_arxiv_id":"2407.21787","is_internal_anchor":true},{"doi":"","year":2021,"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","ref_index":5,"cited_arxiv_id":"2110.14168","is_internal_anchor":true}],"resolved_work":65,"snapshot_sha256":"c641e77b14dc291d581e8353a882323638234b3b5972594db6be58da7cac2ea4","internal_anchors":7},"formal_canon":{"evidence_count":2,"snapshot_sha256":"ccccfa6436f3acdb87894e5442821afb4e80ad60b5f5b3ed0c56ec6090bc8829"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.12620","created_at":"2026-05-18T03:10:00.447563+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.12620v1","created_at":"2026-05-18T03:10:00.447563+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12620","created_at":"2026-05-18T03:10:00.447563+00:00"},{"alias_kind":"pith_short_12","alias_value":"BDDQ3SU5C4FN","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"BDDQ3SU5C4FNMMBT","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"BDDQ3SU5","created_at":"2026-05-18T12:33:37.589309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4","json":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4.json","graph_json":"https://pith.science/api/pith-number/BDDQ3SU5C4FNMMBT2LSZ5F54N4/graph.json","events_json":"https://pith.science/api/pith-number/BDDQ3SU5C4FNMMBT2LSZ5F54N4/events.json","paper":"https://pith.science/paper/BDDQ3SU5"},"agent_actions":{"view_html":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4","download_json":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4.json","view_paper":"https://pith.science/paper/BDDQ3SU5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.12620&json=true","fetch_graph":"https://pith.science/api/pith-number/BDDQ3SU5C4FNMMBT2LSZ5F54N4/graph.json","fetch_events":"https://pith.science/api/pith-number/BDDQ3SU5C4FNMMBT2LSZ5F54N4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4/action/storage_attestation","attest_author":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4/action/author_attestation","sign_citation":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4/action/citation_signature","submit_replication":"https://pith.science/pith/BDDQ3SU5C4FNMMBT2LSZ5F54N4/action/replication_record"}},"created_at":"2026-05-18T03:10:00.447563+00:00","updated_at":"2026-05-18T03:10:00.447563+00:00"}