{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/eagle-2-5-boosting-long-context-post-training","title":"Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models","arxiv_id":"2504.15271","date":"2025-04-21","proceeding":null,"authors":["Guo Chen","Zhiqi Li","Shihao Wang","Jindong Jiang","Yicheng Liu","Lidong Lu","De-An Huang","Wonmin Byeon","Matthieu Le","Max Ehrlich","Tuomas Rintamaki","Tyler Poon","Tong Lu","LiMin Wang","Bryan Catanzaro","Jan Kautz","Andrew Tao","Zhiding Yu","Guilin Liu"],"abstract":"We introduce Eagle 2.5, a family of frontier vision-language models (VLMs) for long-context multimodal learning. Our work addresses the challenges in long video comprehension and high-resolution image understanding, introducing a generalist framework for both tasks. The proposed training framework incorporates Automatic Degrade Sampling and Image Area Preservation, two techniques that preserve contextual integrity and visual details. The framework also includes numerous efficiency optimizations in the pipeline for long-context data training. Finally, we propose Eagle-Video-110K, a novel dataset that integrates both story-level and clip-level annotations, facilitating long-video understanding. Eagle 2.5 demonstrates substantial improvements on long-context multimodal benchmarks, providing a robust solution to the limitations of existing VLMs. Notably, our best model Eagle 2.5-8B achieves 72.4% on Video-MME with 512 input frames, matching the results of top-tier commercial model such as GPT-4o and large-scale open-source models like Qwen2.5-VL-72B and InternVL2.5-78B.","url_abs":"https://arxiv.org/abs/2504.15271v1","url_pdf":"https://arxiv.org/pdf/2504.15271v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"eagle-2-5-boosting-long-context-post-training","repo_url":"https://github.com/nvlabs/eagle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"mme","task_name":"MME"},{"task_slug":null,"task_name":"Video MME"},{"task_slug":"video-understanding","task_name":"Video Understanding"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.15271","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.15271"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nvlabs/eagle","reach":null}],"summary":{"ran_draft_wrong":1,"ran_fixture":2},"by_repo_kind":{"listed":{"samples":1,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"1cebd4910bcf937b","entry":"args_to_build_options","repo":"nvlabs/eagle","repo_kind":"listed","path":"Eagle2_5/deployment/export_llm_engine.py","file_url":"https://github.com/nvlabs/eagle/blob/HEAD/Eagle2_5/deployment/export_llm_engine.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1cebd4910bcf937b"}},{"code_sha256_prefix":"48ca29fb5a02f269","entry":"find_sequences_torch","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"48ca29fb5a02f269"}},{"code_sha256_prefix":"e3afee94fb11f8a9","entry":"pre_calc_loss_weight","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"e3afee94fb11f8a9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}