{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SM5S6LWJOLOE3G2NAESZXD757Q","short_pith_number":"pith:SM5S6LWJ","schema_version":"1.0","canonical_sha256":"933b2f2ec972dc4d9b4d01259b8ffdfc0a40c6ee1d7b02cb3110c9451fc5524f","source":{"kind":"arxiv","id":"2209.12356","version":2},"attestation_state":"computed","paper":{"title":"News Summarization and Evaluation in the Era of GPT-3","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Greg Durrett, Junyi Jessy Li, Tanya Goyal","submitted_at":"2022-09-26T01:04:52Z","abstract_excerpt":"The recent success of prompting large language models like GPT-3 has led to a paradigm shift in NLP research. In this paper, we study its impact on text summarization, focusing on the classic benchmark domain of news summarization. First, we investigate how GPT-3 compares against fine-tuned models trained on large summarization datasets. We show that not only do humans overwhelmingly prefer GPT-3 summaries, prompted using only a task description, but these also do not suffer from common dataset-specific issues such as poor factuality. Next, we study what this means for evaluation, particularly"},"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":"2209.12356","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-09-26T01:04:52Z","cross_cats_sorted":[],"title_canon_sha256":"ff0f4f5ee174c7721c13f0c89d59b076008bd4442c5ea362aef591972ced2fb5","abstract_canon_sha256":"31ea73064087bffd62ae5b725d18e022d7a7829d0091dad5b985fc21b71abb12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:16.529656Z","signature_b64":"B9pGvf6zI5JVKptAs46vn/DUH2vnXpMBgXr2dDVV0rJbYHykA6irLIAbUrS/+5zbXhadKn3SMl2hcodQcd9VCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"933b2f2ec972dc4d9b4d01259b8ffdfc0a40c6ee1d7b02cb3110c9451fc5524f","last_reissued_at":"2026-07-05T06:13:16.529146Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:16.529146Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"News Summarization and Evaluation in the Era of GPT-3","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Greg Durrett, Junyi Jessy Li, Tanya Goyal","submitted_at":"2022-09-26T01:04:52Z","abstract_excerpt":"The recent success of prompting large language models like GPT-3 has led to a paradigm shift in NLP research. In this paper, we study its impact on text summarization, focusing on the classic benchmark domain of news summarization. First, we investigate how GPT-3 compares against fine-tuned models trained on large summarization datasets. We show that not only do humans overwhelmingly prefer GPT-3 summaries, prompted using only a task description, but these also do not suffer from common dataset-specific issues such as poor factuality. Next, we study what this means for evaluation, particularly"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.12356","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/2209.12356/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":"2209.12356","created_at":"2026-07-05T06:13:16.529205+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.12356v2","created_at":"2026-07-05T06:13:16.529205+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.12356","created_at":"2026-07-05T06:13:16.529205+00:00"},{"alias_kind":"pith_short_12","alias_value":"SM5S6LWJOLOE","created_at":"2026-07-05T06:13:16.529205+00:00"},{"alias_kind":"pith_short_16","alias_value":"SM5S6LWJOLOE3G2N","created_at":"2026-07-05T06:13:16.529205+00:00"},{"alias_kind":"pith_short_8","alias_value":"SM5S6LWJ","created_at":"2026-07-05T06:13:16.529205+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":20,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23989","citing_title":"Faithful by Construction: Claim-Anchored Attribution for Multi-Document Summarization","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13111","citing_title":"M\\\"OVE: A Holistic LLM Benchmark for the German Public Sector","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23989","citing_title":"Faithful by Construction: Claim-Anchored Attribution for Multi-Document Summarization","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05494","citing_title":"A Multi-Model Metric-based Selection Framework for Abstractive Text summarization","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01252","citing_title":"Understanding LLM Behavior in Multi-Target Cross-Lingual Summarization","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26154","citing_title":"MemMorph: Tool Hijacking in LLM Agents via Memory Poisoning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05494","citing_title":"A Multi-Model Metric-based Selection Framework for Abstractive Text summarization","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2406.15809","citing_title":"LaMSUM: Amplifying Voices Against Harassment through LLM Guided Extractive Summarization of User Incident Reports","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2412.14368","citing_title":"Beyond Math: Stories as a Testbed for Memorization-Constrained Reasoning in LLMs","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15104","citing_title":"From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents","ref_index":145,"is_internal_anchor":false},{"citing_arxiv_id":"2511.13658","citing_title":"Why is \"Chicago\" Predictive of Deceptive Reviews? Using LLMs to Discover Language Phenomena from Lexical Cues","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2302.04023","citing_title":"A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2409.12917","citing_title":"Training Language Models to Self-Correct via Reinforcement Learning","ref_index":107,"is_internal_anchor":false},{"citing_arxiv_id":"2309.00267","citing_title":"RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2303.17564","citing_title":"BloombergGPT: A Large Language Model for Finance","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12227","citing_title":"A Recipe for Long-Context Reasoning in Large Language Models via On-Policy Optimization and Distillation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26590","citing_title":"Recommendations for Efficient and Responsible LLM Adoption within Industrial Software Development","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25665","citing_title":"LLM-ReSum: A Framework for LLM Reflective Summarization through Self-Evaluation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10883","citing_title":"Towards Designing for Resilience: Community-Centered Deployment of an AI Business Planning Tool in a Small Business Center","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02300","citing_title":"A Meta Reinforcement Learning Approach to Goals-Based Wealth Management","ref_index":297,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q","json":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q.json","graph_json":"https://pith.science/api/pith-number/SM5S6LWJOLOE3G2NAESZXD757Q/graph.json","events_json":"https://pith.science/api/pith-number/SM5S6LWJOLOE3G2NAESZXD757Q/events.json","paper":"https://pith.science/paper/SM5S6LWJ"},"agent_actions":{"view_html":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q","download_json":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q.json","view_paper":"https://pith.science/paper/SM5S6LWJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.12356&json=true","fetch_graph":"https://pith.science/api/pith-number/SM5S6LWJOLOE3G2NAESZXD757Q/graph.json","fetch_events":"https://pith.science/api/pith-number/SM5S6LWJOLOE3G2NAESZXD757Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q/action/storage_attestation","attest_author":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q/action/author_attestation","sign_citation":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q/action/citation_signature","submit_replication":"https://pith.science/pith/SM5S6LWJOLOE3G2NAESZXD757Q/action/replication_record"}},"created_at":"2026-07-05T06:13:16.529205+00:00","updated_at":"2026-07-05T06:13:16.529205+00:00"}