{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ZDDH5CEBC3WYDZV7J3DU2G7DFT","short_pith_number":"pith:ZDDH5CEB","canonical_record":{"source":{"id":"2606.31270","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T07:44:37Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CY","cs.LG"],"title_canon_sha256":"d67b2af9867465e9d3823b3782d1f702fc7382d80dae0f2ed50622867c953137","abstract_canon_sha256":"90673e347d1102a3c4adc0fbab7d0d1006d608216a4afab56467579e949cf077"},"schema_version":"1.0"},"canonical_sha256":"c8c67e888116ed81e6bf4ec74d1be32cda95bb08b7cf17e89cd12c848c625006","source":{"kind":"arxiv","id":"2606.31270","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.31270","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"arxiv_version","alias_value":"2606.31270v1","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31270","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_12","alias_value":"ZDDH5CEBC3WY","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_16","alias_value":"ZDDH5CEBC3WYDZV7","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_8","alias_value":"ZDDH5CEB","created_at":"2026-07-01T01:17:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ZDDH5CEBC3WYDZV7J3DU2G7DFT","target":"record","payload":{"canonical_record":{"source":{"id":"2606.31270","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T07:44:37Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CY","cs.LG"],"title_canon_sha256":"d67b2af9867465e9d3823b3782d1f702fc7382d80dae0f2ed50622867c953137","abstract_canon_sha256":"90673e347d1102a3c4adc0fbab7d0d1006d608216a4afab56467579e949cf077"},"schema_version":"1.0"},"canonical_sha256":"c8c67e888116ed81e6bf4ec74d1be32cda95bb08b7cf17e89cd12c848c625006","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T01:17:57.680655Z","signature_b64":"xDZ8rX4tNycJ9dKKb7f/6QKwBUTwgQSjzbe0PND+lCTthijyrzH0o5KSaP6NR+rfeM9GRaRjF6QMjEkXqBE6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8c67e888116ed81e6bf4ec74d1be32cda95bb08b7cf17e89cd12c848c625006","last_reissued_at":"2026-07-01T01:17:57.680220Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T01:17:57.680220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.31270","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-01T01:17:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zkSknCAEiV7CPxXQIaqqSVxdBQm+KHozQRLFAlI/jczAVY9bP1vgRN4G5sK/0WzNXyY9P8gei00/s+4AAbzuDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-01T19:14:34.701554Z"},"content_sha256":"0cf3e9e9b068dee7c62c19774a9341b108974747442613acedf903e3c91bbeba","schema_version":"1.0","event_id":"sha256:0cf3e9e9b068dee7c62c19774a9341b108974747442613acedf903e3c91bbeba"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ZDDH5CEBC3WYDZV7J3DU2G7DFT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CY","cs.LG"],"primary_cat":"cs.CV","authors_text":"Ludwig Schmidt, Serena Yeung-Levy, Xiaohan Wang, Xueqiao Sun, Yuhui Zhang","submitted_at":"2026-06-30T07:44:37Z","abstract_excerpt":"Computer-use agents, which leverage multimodal large language models (MLLMs) to operate computers and complete tasks, have attracted significant attention for their utility and versatility. A major challenge in developing these agents is collecting large-scale, high-quality trajectories. The standard approach generates synthetic data through a self-improving loop: an agent is placed in a verifiable environment and iteratively fine-tuned on its successful trajectories. Despite its effectiveness, this paradigm exploits only successful trajectories and discards the failed ones, even though failur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31270","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.31270/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-01T01:17:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zJAAr7Gj/TnTSJtUyGr8I5vylT1/GjA2bW3MgnYtVgUCl3XIC9YKj5AESMT/s6RO6c3LKrjd/vG6YQXNshm8Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-01T19:14:34.701921Z"},"content_sha256":"c64ad9d80b9a38336fe296175f3e250d0247f491c432604972a1427ff497c938","schema_version":"1.0","event_id":"sha256:c64ad9d80b9a38336fe296175f3e250d0247f491c432604972a1427ff497c938"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/bundle.json","state_url":"https://pith.science/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-01T19:14:34Z","links":{"resolver":"https://pith.science/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT","bundle":"https://pith.science/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/bundle.json","state":"https://pith.science/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZDDH5CEBC3WYDZV7J3DU2G7DFT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZDDH5CEBC3WYDZV7J3DU2G7DFT","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"90673e347d1102a3c4adc0fbab7d0d1006d608216a4afab56467579e949cf077","cross_cats_sorted":["cs.AI","cs.CL","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T07:44:37Z","title_canon_sha256":"d67b2af9867465e9d3823b3782d1f702fc7382d80dae0f2ed50622867c953137"},"schema_version":"1.0","source":{"id":"2606.31270","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.31270","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"arxiv_version","alias_value":"2606.31270v1","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31270","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_12","alias_value":"ZDDH5CEBC3WY","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_16","alias_value":"ZDDH5CEBC3WYDZV7","created_at":"2026-07-01T01:17:57Z"},{"alias_kind":"pith_short_8","alias_value":"ZDDH5CEB","created_at":"2026-07-01T01:17:57Z"}],"graph_snapshots":[{"event_id":"sha256:c64ad9d80b9a38336fe296175f3e250d0247f491c432604972a1427ff497c938","target":"graph","created_at":"2026-07-01T01:17:57Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.31270/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Computer-use agents, which leverage multimodal large language models (MLLMs) to operate computers and complete tasks, have attracted significant attention for their utility and versatility. A major challenge in developing these agents is collecting large-scale, high-quality trajectories. The standard approach generates synthetic data through a self-improving loop: an agent is placed in a verifiable environment and iteratively fine-tuned on its successful trajectories. Despite its effectiveness, this paradigm exploits only successful trajectories and discards the failed ones, even though failur","authors_text":"Ludwig Schmidt, Serena Yeung-Levy, Xiaohan Wang, Xueqiao Sun, Yuhui Zhang","cross_cats":["cs.AI","cs.CL","cs.CY","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T07:44:37Z","title":"Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31270","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0cf3e9e9b068dee7c62c19774a9341b108974747442613acedf903e3c91bbeba","target":"record","created_at":"2026-07-01T01:17:57Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"90673e347d1102a3c4adc0fbab7d0d1006d608216a4afab56467579e949cf077","cross_cats_sorted":["cs.AI","cs.CL","cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T07:44:37Z","title_canon_sha256":"d67b2af9867465e9d3823b3782d1f702fc7382d80dae0f2ed50622867c953137"},"schema_version":"1.0","source":{"id":"2606.31270","kind":"arxiv","version":1}},"canonical_sha256":"c8c67e888116ed81e6bf4ec74d1be32cda95bb08b7cf17e89cd12c848c625006","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c8c67e888116ed81e6bf4ec74d1be32cda95bb08b7cf17e89cd12c848c625006","first_computed_at":"2026-07-01T01:17:57.680220Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T01:17:57.680220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xDZ8rX4tNycJ9dKKb7f/6QKwBUTwgQSjzbe0PND+lCTthijyrzH0o5KSaP6NR+rfeM9GRaRjF6QMjEkXqBE6CA==","signature_status":"signed_v1","signed_at":"2026-07-01T01:17:57.680655Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.31270","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0cf3e9e9b068dee7c62c19774a9341b108974747442613acedf903e3c91bbeba","sha256:c64ad9d80b9a38336fe296175f3e250d0247f491c432604972a1427ff497c938"],"state_sha256":"4fb87f480bbad2d25888fce76d654767468b34eea2e20f89a519304fb4ae37ef"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9C+E52xpr/dpYj6HR3S/qa6DpuQPquUWmEvfrpwU9mEgRKL8UHUQl9MoRqFoEsVMBSNCuNSOCm7zxIjXWMxICg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-01T19:14:34.703818Z","bundle_sha256":"7335f4def6730caa4cbe5c5d5d904eab9d38afe7128dbd45033f5d54d7b67d0a"}}