{"work":{"id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","openalex_id":"https://openalex.org/W4414359541","doi":"10.24963/ijcai.2025/706","arxiv_id":"2307.09288","raw_key":null,"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","authors":null,"authors_text":"Hugo Touvron, Louis Martin, Kevin Stone, et al","year":2023,"venue":"cs.CL","abstract":"In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on our work and contribute to the responsible development of LLMs.","external_url":"https://arxiv.org/abs/2307.09288","cited_by_count":0,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2307.09288","created_at":"2026-05-08T18:13:54.069555+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","render_title":"Llama 2: Open Foundation and Fine-Tuned Chat Models"},"hub":{"state":{"work_id":"68a5177f-d644-44c1-bd4f-4e5278c22f5d","tier":"mega_hub","tier_reason":"1,000+ Pith inbound or 100,000+ external citations","pith_inbound_count":1105,"external_cited_by_count":0,"distinct_field_count":37,"first_pith_cited_at":"2023-02-03T06:06:27+00:00","last_pith_cited_at":"2026-07-09T12:28:39+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"needed","recognition_status":"needed","updated_at":"2026-08-20T17:59:45.053513+00:00","tier_text":"mega_hub"},"tier":"mega_hub","role_counts":[{"context_role":"background","n":154},{"context_role":"method","n":21},{"context_role":"dataset","n":9},{"context_role":"baseline","n":6},{"context_role":"other","n":3},{"context_role":"extension","n":1}],"polarity_counts":[{"context_polarity":"background","n":148},{"context_polarity":"use_method","n":21},{"context_polarity":"use_dataset","n":9},{"context_polarity":"unclear","n":7},{"context_polarity":"baseline","n":6},{"context_polarity":"support","n":2},{"context_polarity":"extend","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","claims":[{"claim_text":"In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on o","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Llama 2: Open Foundation and Fine-Tuned Chat Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T18:43:30.020349+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"6af248bf-4977-4496-b743-94d90183b7a5","orcid":null,"display_name":"Hugo Touvron"},{"id":"1fe8ecee-1c46-42b9-87fb-3b160972fb65","orcid":null,"display_name":"Louis Martin"},{"id":"643d0963-d7dd-4acc-ad85-741c731071d1","orcid":null,"display_name":"Kevin Stone"},{"id":"535def4b-a2c7-4c75-ad8b-db3e939f2bc9","orcid":null,"display_name":"et al"}]},"error":null,"updated_at":"2026-05-13T18:43:30.018391+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-13T18:43:29.640696+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":111},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":95},{"title":"Mistral 7B","work_id":"eb5e1305-ad11-4875-ad8d-ad8b8f697599","shared_citers":62},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":54},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":52},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":51},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":49},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":42},{"title":"Gemini: A Family of Highly Capable Multimodal Models","work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","shared_citers":40},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":40},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":31},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":30},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":29},{"title":"Code Llama: Open Foundation Models for Code","work_id":"e73bffa4-7620-47ac-9327-259a60db52ca","shared_citers":25},{"title":"Gemma: Open Models Based on Gemini Research and Technology","work_id":"a9ea2870-df28-40b8-a9e0-a7e9a116f793","shared_citers":25},{"title":"Universal and Transferable Adversarial Attacks on Aligned Language Models","work_id":"3322fa86-1768-4677-8425-dd326b45e078","shared_citers":25},{"title":"Qwen2 Technical Report","work_id":"a1857881-ab9b-4b80-9b5f-9ae4b5c2566d","shared_citers":24},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":23},{"title":"Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback","work_id":"a1f2574b-a899-4713-be60-c87ba332656c","shared_citers":23},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":22},{"title":"Measuring Massive Multitask Language Understanding","work_id":"e87ec49a-544b-4ec8-8991-75298c64ff5e","shared_citers":22},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":22},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":21},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":20}],"time_series":[{"n":19,"year":2023},{"n":39,"year":2024},{"n":7,"year":2025},{"n":274,"year":2026}]},"error":null,"updated_at":"2026-05-13T17:25:56.537419+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"fixed":1,"items":[{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-13T18:43:29.014608+00:00"},"reader_index":{"job_type":"reader_index","status":"succeeded","result":{"note":"annotated reader requires full-text/OA fetch; shell is wired for mega hubs","status":"reader queued"},"error":null,"updated_at":"2026-07-03T00:33:14.325722+00:00"},"recognition_alignment":{"job_type":"recognition_alignment","status":"succeeded","result":{"modules":["IndisputableMonolith.Chemistry.VanDerWaals","IndisputableMonolith.Cosmology.DarkEnergy","IndisputableMonolith.Physics.GrandUnificationFromRS","IndisputableMonolith.StandardModel.StrongCP","IndisputableMonolith.Linguistics.PhonemeInventoryBandFromRS","IndisputableMonolith.QFT.ElectroweakScaleStructure","IndisputableMonolith.Sociology.DunbarFromBandwidth","IndisputableMonolith.Foundation.AlexanderDuality"],"query_chars":714},"error":null,"updated_at":"2026-07-03T00:33:34.008194+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","claims":[{"claim_text":"In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on o","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Llama 2: Open Foundation and Fine-Tuned Chat Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T18:43:29.644957+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","claims":[{"claim_text":"In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on o","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Llama 2: Open Foundation and Fine-Tuned Chat Models because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T17:25:52.728519+00:00"}},"summary":{"title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","claims":[{"claim_text":"In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on o","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Llama 2: Open Foundation and Fine-Tuned Chat Models because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","shared_citers":111},{"title":"The Llama 3 Herd of Models","work_id":"1549a635-88af-4ac1-acfe-51ae7bb53345","shared_citers":95},{"title":"Mistral 7B","work_id":"eb5e1305-ad11-4875-ad8d-ad8b8f697599","shared_citers":62},{"title":"Training Verifiers to Solve Math Word Problems","work_id":"acab1aa8-b4d6-40e0-a3ee-25341701dca2","shared_citers":54},{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","shared_citers":52},{"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","shared_citers":51},{"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","shared_citers":49},{"title":"Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge","work_id":"28ea1282-d657-4c61-a83c-f1249be6d6b1","shared_citers":42},{"title":"Gemini: A Family of Highly Capable Multimodal Models","work_id":"83f7c85b-3f11-450f-ac0c-64d9745220b2","shared_citers":40},{"title":"Qwen Technical Report","work_id":"bb1fd52f-6b2f-437c-9516-37bdf6eb9be8","shared_citers":40},{"title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","work_id":"e6b75ad5-2877-4168-97c8-710407094d20","shared_citers":31},{"title":"Qwen2.5 Technical Report","work_id":"d8432992-4980-4a81-85c7-9fa2c2b87f85","shared_citers":30},{"title":"Scaling Laws for Neural Language Models","work_id":"b7dd8749-9c45-4977-ab9b-64478dce1ae8","shared_citers":29},{"title":"Code Llama: Open Foundation Models for Code","work_id":"e73bffa4-7620-47ac-9327-259a60db52ca","shared_citers":25},{"title":"Gemma: Open Models Based on Gemini Research and Technology","work_id":"a9ea2870-df28-40b8-a9e0-a7e9a116f793","shared_citers":25},{"title":"Universal and Transferable Adversarial Attacks on Aligned Language Models","work_id":"3322fa86-1768-4677-8425-dd326b45e078","shared_citers":25},{"title":"Qwen2 Technical Report","work_id":"a1857881-ab9b-4b80-9b5f-9ae4b5c2566d","shared_citers":24},{"title":"DeepSeek-V3 Technical Report","work_id":"57d2791d-2219-4c31-a077-afc04b12a75c","shared_citers":23},{"title":"Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback","work_id":"a1f2574b-a899-4713-be60-c87ba332656c","shared_citers":23},{"title":"Constitutional AI: Harmlessness from AI Feedback","work_id":"faaaa4e0-2676-4fac-a0b4-99aef10d2095","shared_citers":22},{"title":"Measuring Massive Multitask Language Understanding","work_id":"e87ec49a-544b-4ec8-8991-75298c64ff5e","shared_citers":22},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":22},{"title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","work_id":"ed240a10-5b19-406c-baa5-30803f465785","shared_citers":21},{"title":"Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities","work_id":"008df105-2fdd-45d8-857a-8e35868aecb6","shared_citers":20}],"time_series":[{"n":19,"year":2023},{"n":39,"year":2024},{"n":7,"year":2025},{"n":274,"year":2026}]},"authors":[{"id":"535def4b-a2c7-4c75-ad8b-db3e939f2bc9","orcid":null,"display_name":"et al","source":"manual","import_confidence":0.72},{"id":"6af248bf-4977-4496-b743-94d90183b7a5","orcid":null,"display_name":"Hugo Touvron","source":"manual","import_confidence":0.72},{"id":"643d0963-d7dd-4acc-ad85-741c731071d1","orcid":null,"display_name":"Kevin Stone","source":"manual","import_confidence":0.72},{"id":"1fe8ecee-1c46-42b9-87fb-3b160972fb65","orcid":null,"display_name":"Louis Martin","source":"manual","import_confidence":0.72}]}}