{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GZI2H24MTVNAB3OEVL2KILS3A5","short_pith_number":"pith:GZI2H24M","canonical_record":{"source":{"id":"2407.15711","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-22T15:18:45Z","cross_cats_sorted":[],"title_canon_sha256":"b1c93856b9b7bd3e00373c9a325faaadb236d7b1954759712ad831d3325ecb21","abstract_canon_sha256":"963e7659fd710b8a513b95dc8753592ae788a345010387637c4c7f710ee92983"},"schema_version":"1.0"},"canonical_sha256":"3651a3eb8c9d5a00edc4aaf4a42e5b077ec3e90dc451b182e7fdc3c873776656","source":{"kind":"arxiv","id":"2407.15711","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.15711","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"arxiv_version","alias_value":"2407.15711v2","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.15711","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_12","alias_value":"GZI2H24MTVNA","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_16","alias_value":"GZI2H24MTVNAB3OE","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_8","alias_value":"GZI2H24M","created_at":"2026-07-05T09:23:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GZI2H24MTVNAB3OEVL2KILS3A5","target":"record","payload":{"canonical_record":{"source":{"id":"2407.15711","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-22T15:18:45Z","cross_cats_sorted":[],"title_canon_sha256":"b1c93856b9b7bd3e00373c9a325faaadb236d7b1954759712ad831d3325ecb21","abstract_canon_sha256":"963e7659fd710b8a513b95dc8753592ae788a345010387637c4c7f710ee92983"},"schema_version":"1.0"},"canonical_sha256":"3651a3eb8c9d5a00edc4aaf4a42e5b077ec3e90dc451b182e7fdc3c873776656","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:23:09.100970Z","signature_b64":"Ka3xTQfDp+CX+Df51scQbBF7LZI3tzcAcvWs/1iwUu5xjjRWmw9i21En8gYOE2FdwPNaRcMzqNyZsHG1IE69DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3651a3eb8c9d5a00edc4aaf4a42e5b077ec3e90dc451b182e7fdc3c873776656","last_reissued_at":"2026-07-05T09:23:09.100521Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:23:09.100521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.15711","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:23:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5l2cGcQZyi8pbm1tlV2mCZ+qd81PjF8l7O6LdiVKXWWnKsCvuAMuuR6X97pLbgZL+HQxHKRgSlfSiEbuzNq0Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:55:19.200451Z"},"content_sha256":"15eab2443131c4f53fc674b4bf11bfc1f96c4fb69f769cd21573426696abcceb","schema_version":"1.0","event_id":"sha256:15eab2443131c4f53fc674b4bf11bfc1f96c4fb69f769cd21573426696abcceb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GZI2H24MTVNAB3OEVL2KILS3A5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ben Bogin, Chaitanya Malaviya, Jonathan Berant, Ofir Press, Ori Yoran, Samuel Joseph Amouyal","submitted_at":"2024-07-22T15:18:45Z","abstract_excerpt":"Language agents, built on top of language models (LMs), are systems that can interact with complex environments, such as the open web. In this work, we examine whether such agents can perform realistic and time-consuming tasks on the web, e.g., monitoring real-estate markets or locating relevant nearby businesses. We introduce AssistantBench, a challenging new benchmark consisting of 214 realistic tasks that can be automatically evaluated, covering different scenarios and domains. We find that AssistantBench exposes the limitations of current systems, including language models and retrieval-au"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.15711","kind":"arxiv","version":2},"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/2407.15711/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":"2026-07-05T09:23:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KOb9GhFeRG31ukZl1DyDWf3UbGKXEq4kPgKVV6ycM589E/IXzY0iX6GohmF1ArjN1XpJMctdJmLtXiLzLOqvAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:55:19.200828Z"},"content_sha256":"5365405ef49b2112ad8f557c0cc639715f1eb18355872c4e5f701033f8708bb7","schema_version":"1.0","event_id":"sha256:5365405ef49b2112ad8f557c0cc639715f1eb18355872c4e5f701033f8708bb7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GZI2H24MTVNAB3OEVL2KILS3A5/bundle.json","state_url":"https://pith.science/pith/GZI2H24MTVNAB3OEVL2KILS3A5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GZI2H24MTVNAB3OEVL2KILS3A5/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":"2026-07-07T06:55:19Z","links":{"resolver":"https://pith.science/pith/GZI2H24MTVNAB3OEVL2KILS3A5","bundle":"https://pith.science/pith/GZI2H24MTVNAB3OEVL2KILS3A5/bundle.json","state":"https://pith.science/pith/GZI2H24MTVNAB3OEVL2KILS3A5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GZI2H24MTVNAB3OEVL2KILS3A5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GZI2H24MTVNAB3OEVL2KILS3A5","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":"963e7659fd710b8a513b95dc8753592ae788a345010387637c4c7f710ee92983","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-22T15:18:45Z","title_canon_sha256":"b1c93856b9b7bd3e00373c9a325faaadb236d7b1954759712ad831d3325ecb21"},"schema_version":"1.0","source":{"id":"2407.15711","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.15711","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"arxiv_version","alias_value":"2407.15711v2","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.15711","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_12","alias_value":"GZI2H24MTVNA","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_16","alias_value":"GZI2H24MTVNAB3OE","created_at":"2026-07-05T09:23:09Z"},{"alias_kind":"pith_short_8","alias_value":"GZI2H24M","created_at":"2026-07-05T09:23:09Z"}],"graph_snapshots":[{"event_id":"sha256:5365405ef49b2112ad8f557c0cc639715f1eb18355872c4e5f701033f8708bb7","target":"graph","created_at":"2026-07-05T09:23:09Z","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/2407.15711/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language agents, built on top of language models (LMs), are systems that can interact with complex environments, such as the open web. In this work, we examine whether such agents can perform realistic and time-consuming tasks on the web, e.g., monitoring real-estate markets or locating relevant nearby businesses. We introduce AssistantBench, a challenging new benchmark consisting of 214 realistic tasks that can be automatically evaluated, covering different scenarios and domains. We find that AssistantBench exposes the limitations of current systems, including language models and retrieval-au","authors_text":"Ben Bogin, Chaitanya Malaviya, Jonathan Berant, Ofir Press, Ori Yoran, Samuel Joseph Amouyal","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-22T15:18:45Z","title":"AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.15711","kind":"arxiv","version":2},"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":"sha256:15eab2443131c4f53fc674b4bf11bfc1f96c4fb69f769cd21573426696abcceb","target":"record","created_at":"2026-07-05T09:23:09Z","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":"963e7659fd710b8a513b95dc8753592ae788a345010387637c4c7f710ee92983","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-22T15:18:45Z","title_canon_sha256":"b1c93856b9b7bd3e00373c9a325faaadb236d7b1954759712ad831d3325ecb21"},"schema_version":"1.0","source":{"id":"2407.15711","kind":"arxiv","version":2}},"canonical_sha256":"3651a3eb8c9d5a00edc4aaf4a42e5b077ec3e90dc451b182e7fdc3c873776656","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3651a3eb8c9d5a00edc4aaf4a42e5b077ec3e90dc451b182e7fdc3c873776656","first_computed_at":"2026-07-05T09:23:09.100521Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:23:09.100521Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ka3xTQfDp+CX+Df51scQbBF7LZI3tzcAcvWs/1iwUu5xjjRWmw9i21En8gYOE2FdwPNaRcMzqNyZsHG1IE69DA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:23:09.100970Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.15711","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:15eab2443131c4f53fc674b4bf11bfc1f96c4fb69f769cd21573426696abcceb","sha256:5365405ef49b2112ad8f557c0cc639715f1eb18355872c4e5f701033f8708bb7"],"state_sha256":"a3fee8e51f7b5b852f1c8671726181d7211e2b95c3d8ae006092b986ef5d036c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H4dM6JcYNinCIo+7teluVdXf3OqHvZuA1YBhdDTO+oLggjFOrr3rkIwu8/pTMqchbXhfwLWZYzdVe7TzwemACA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:55:19.202818Z","bundle_sha256":"12bbcdafb87615c57a31bb0d1464240c02c50b7a9c383aafb99eccc13e706433"}}