{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:M6QLUGOGIF7RR3KH6JW6LAXWCC","short_pith_number":"pith:M6QLUGOG","schema_version":"1.0","canonical_sha256":"67a0ba19c6417f18ed47f26de582f610be0c2b4e2a52624942e09501bf41dda4","source":{"kind":"arxiv","id":"2310.15747","version":2},"attestation_state":"computed","paper":{"title":"Large Language Models are Temporal and Causal Reasoners for Video Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Byungseok Roh, Dohwan Ko, Hyunwoo J. Kim, Ji Soo Lee, Wooyoung Kang","submitted_at":"2023-10-24T11:44:39Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable performances on a wide range of natural language understanding and generation tasks. We observe that the LLMs provide effective priors in exploiting $\\textit{linguistic shortcuts}$ for temporal and causal reasoning in Video Question Answering (VideoQA). However, such priors often cause suboptimal results on VideoQA by leading the model to over-rely on questions, $\\textit{i.e.}$, $\\textit{linguistic bias}$, while ignoring visual content. This is also known as `ungrounded guesses' or `hallucinations'. To address this problem while leveraging LLM"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2310.15747","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-10-24T11:44:39Z","cross_cats_sorted":[],"title_canon_sha256":"ba52d933a1c1edb0023b0c4c1bfe22e14a9e1c23fab1f9d19fe20a17c7c8d7c5","abstract_canon_sha256":"e381858ac63dbb12e2926f1d19e14a4f34b026ed611bc5bb8fda9c50632abbbb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:09:32.664437Z","signature_b64":"ENNUwmQvuBwzxQXlVhRITFR5k8fVegcCvW+83vzSludR5IUWz4TSHoXzJYA6JTeEWqqPgq/Ywmv/J+7csZS1CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67a0ba19c6417f18ed47f26de582f610be0c2b4e2a52624942e09501bf41dda4","last_reissued_at":"2026-07-05T07:09:32.663949Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:09:32.663949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Language Models are Temporal and Causal Reasoners for Video Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Byungseok Roh, Dohwan Ko, Hyunwoo J. Kim, Ji Soo Lee, Wooyoung Kang","submitted_at":"2023-10-24T11:44:39Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable performances on a wide range of natural language understanding and generation tasks. We observe that the LLMs provide effective priors in exploiting $\\textit{linguistic shortcuts}$ for temporal and causal reasoning in Video Question Answering (VideoQA). However, such priors often cause suboptimal results on VideoQA by leading the model to over-rely on questions, $\\textit{i.e.}$, $\\textit{linguistic bias}$, while ignoring visual content. This is also known as `ungrounded guesses' or `hallucinations'. To address this problem while leveraging LLM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.15747","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/2310.15747/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2310.15747","created_at":"2026-07-05T07:09:32.664003+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.15747v2","created_at":"2026-07-05T07:09:32.664003+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.15747","created_at":"2026-07-05T07:09:32.664003+00:00"},{"alias_kind":"pith_short_12","alias_value":"M6QLUGOGIF7R","created_at":"2026-07-05T07:09:32.664003+00:00"},{"alias_kind":"pith_short_16","alias_value":"M6QLUGOGIF7RR3KH","created_at":"2026-07-05T07:09:32.664003+00:00"},{"alias_kind":"pith_short_8","alias_value":"M6QLUGOG","created_at":"2026-07-05T07:09:32.664003+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17798","citing_title":"LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06991","citing_title":"Don't Pause: Streaming Video-Language Synchrony for Online Video Understanding","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07180","citing_title":"Flattery in Motion: Benchmarking and Analyzing Sycophancy in Video-LLMs","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC","json":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC.json","graph_json":"https://pith.science/api/pith-number/M6QLUGOGIF7RR3KH6JW6LAXWCC/graph.json","events_json":"https://pith.science/api/pith-number/M6QLUGOGIF7RR3KH6JW6LAXWCC/events.json","paper":"https://pith.science/paper/M6QLUGOG"},"agent_actions":{"view_html":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC","download_json":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC.json","view_paper":"https://pith.science/paper/M6QLUGOG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.15747&json=true","fetch_graph":"https://pith.science/api/pith-number/M6QLUGOGIF7RR3KH6JW6LAXWCC/graph.json","fetch_events":"https://pith.science/api/pith-number/M6QLUGOGIF7RR3KH6JW6LAXWCC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC/action/storage_attestation","attest_author":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC/action/author_attestation","sign_citation":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC/action/citation_signature","submit_replication":"https://pith.science/pith/M6QLUGOGIF7RR3KH6JW6LAXWCC/action/replication_record"}},"created_at":"2026-07-05T07:09:32.664003+00:00","updated_at":"2026-07-05T07:09:32.664003+00:00"}