{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:32GMPYZPDA6GC4M2MVJO7WOGEP","short_pith_number":"pith:32GMPYZP","schema_version":"1.0","canonical_sha256":"de8cc7e32f183c61719a6552efd9c623ddf44e989df98db79dd131d58f012e29","source":{"kind":"arxiv","id":"2107.08430","version":2},"attestation_state":"computed","paper":{"title":"YOLOX: Exceeding YOLO Series in 2021","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"YOLOX turns YOLO detectors anchor-free with a decoupled head and SimOTA assignment to reach higher accuracy at real-time speeds.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Wang, Jian Sun, Songtao Liu, Zeming Li, Zheng Ge","submitted_at":"2021-07-18T12:55:11Z","abstract_excerpt":"In this report, we present some experienced improvements to YOLO series, forming a new high-performance detector -- YOLOX. We switch the YOLO detector to an anchor-free manner and conduct other advanced detection techniques, i.e., a decoupled head and the leading label assignment strategy SimOTA to achieve state-of-the-art results across a large scale range of models: For YOLO-Nano with only 0.91M parameters and 1.08G FLOPs, we get 25.3% AP on COCO, surpassing NanoDet by 1.8% AP; for YOLOv3, one of the most widely used detectors in industry, we boost it to 47.3% AP on COCO, outperforming the c"},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2107.08430","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-07-18T12:55:11Z","cross_cats_sorted":[],"title_canon_sha256":"448e0cfa55f7964fea285d0d52fd7c8cb392d9b8f8b2cafca786226b493e7440","abstract_canon_sha256":"28db477965163b83d1fc4b948c58e416a5eee1c6cb06987ee4131561b1ef4d81"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:03:41.413898Z","signature_b64":"0pPVJVxieNOoC0z3xfiHy30bWbcoL/oy9qM8tMCAonVZkXx0gsbcf4N5WvmNGuHF3GSwEvoidOrudmnxwoiOAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de8cc7e32f183c61719a6552efd9c623ddf44e989df98db79dd131d58f012e29","last_reissued_at":"2026-07-05T03:03:41.413431Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:03:41.413431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"YOLOX: Exceeding YOLO Series in 2021","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"YOLOX turns YOLO detectors anchor-free with a decoupled head and SimOTA assignment to reach higher accuracy at real-time speeds.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Wang, Jian Sun, Songtao Liu, Zeming Li, Zheng Ge","submitted_at":"2021-07-18T12:55:11Z","abstract_excerpt":"In this report, we present some experienced improvements to YOLO series, forming a new high-performance detector -- YOLOX. We switch the YOLO detector to an anchor-free manner and conduct other advanced detection techniques, i.e., a decoupled head and the leading label assignment strategy SimOTA to achieve state-of-the-art results across a large scale range of models: For YOLO-Nano with only 0.91M parameters and 1.08G FLOPs, we get 25.3% AP on COCO, surpassing NanoDet by 1.8% AP; for YOLOv3, one of the most widely used detectors in industry, we boost it to 47.3% AP on COCO, outperforming the c"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"for YOLOX-L with roughly the same amount of parameters as YOLOv4-CSP, YOLOv5-L, we achieve 50.0% AP on COCO at a speed of 68.9 FPS on Tesla V100, exceeding YOLOv5-L by 1.8% AP.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the reported gains stem primarily from the anchor-free design, decoupled head, and SimOTA rather than from differences in training schedules, data augmentation, or hyperparameter choices versus the YOLOv5 and YOLOv4 baselines.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"YOLOX exceeds prior YOLO models by adopting anchor-free detection, decoupled heads, and SimOTA assignment to reach 50.0% AP on COCO for the large variant.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"YOLOX turns YOLO detectors anchor-free with a decoupled head and SimOTA assignment to reach higher accuracy at real-time speeds.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"23875a63c2afbdd499b8d891ea62ddd793d594628ffe4885a262a1f9e1dbdd6f"},"source":{"id":"2107.08430","kind":"arxiv","version":2},"verdict":{"id":"b98d8262-d0ae-4388-bb8b-7d4390baf489","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T10:27:58.064967Z","strongest_claim":"for YOLOX-L with roughly the same amount of parameters as YOLOv4-CSP, YOLOv5-L, we achieve 50.0% AP on COCO at a speed of 68.9 FPS on Tesla V100, exceeding YOLOv5-L by 1.8% AP.","one_line_summary":"YOLOX exceeds prior YOLO models by adopting anchor-free detection, decoupled heads, and SimOTA assignment to reach 50.0% AP on COCO for the large variant.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the reported gains stem primarily from the anchor-free design, decoupled head, and SimOTA rather than from differences in training schedules, data augmentation, or hyperparameter choices versus the YOLOv5 and YOLOv4 baselines.","pith_extraction_headline":"YOLOX turns YOLO detectors anchor-free with a decoupled head and SimOTA assignment to reach higher accuracy at real-time speeds."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2107.08430/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":40,"sample":[{"doi":"","year":2004,"title":"YOLOv4: Optimal Speed and Accuracy of Object Detection","work_id":"7057aaee-27f6-4209-a83c-f59727f937a8","ref_index":1,"cited_arxiv_id":"2004.10934","is_internal_anchor":true},{"doi":"","year":2020,"title":"End-to- end object detection with transformers","work_id":"554f512b-c3f2-430c-8d21-0cc7705ef850","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"You only look one-level feature","work_id":"4edbb038-a2ae-4c70-9133-3a04bb2282f1","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Ota: Optimal transport assignment for object detection","work_id":"bd871ef1-d3b6-4e0d-be6a-cc48768ac5bc","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Lla: Loss-aware label assignment for dense pedestrian detection","work_id":"b28e1fc5-2165-4c58-aca9-23ff38edc682","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":40,"snapshot_sha256":"3ed54aa125adb18e7fd94b7e036da8d89e584eed1be7e5cbf4549ebf0d233eba","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"e9064f718d6c87deee7ddec859d99343e86d2c7f16064d178b750d9baefbd470"},"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":"2107.08430","created_at":"2026-07-05T03:03:41.413493+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.08430v2","created_at":"2026-07-05T03:03:41.413493+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.08430","created_at":"2026-07-05T03:03:41.413493+00:00"},{"alias_kind":"pith_short_12","alias_value":"32GMPYZPDA6G","created_at":"2026-07-05T03:03:41.413493+00:00"},{"alias_kind":"pith_short_16","alias_value":"32GMPYZPDA6GC4M2","created_at":"2026-07-05T03:03:41.413493+00:00"},{"alias_kind":"pith_short_8","alias_value":"32GMPYZP","created_at":"2026-07-05T03:03:41.413493+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":65,"internal_anchor_count":65,"sample":[{"citing_arxiv_id":"2606.25312","citing_title":"LEVIRDet: A Million-Scale 159-Category Dataset and Foundation Model for Universal Remote Sensing Object Detection","ref_index":73,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23604","citing_title":"Polycepta: Object-Centric Appearance Estimation for Multi-Object Tracking","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23825","citing_title":"From Spatial to Spectral: An Efficient, Frequency-Guided Feature Representation Learner for Small Object Detection","ref_index":96,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22682","citing_title":"Integrated cloud-based architecture for robot-robot and human-robot collaboration using ROS 2--MQTT in Mediterranean Greenhouses","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19835","citing_title":"Neural Events: Discrete Asynchronous Autoencoders for Event-Based Vision","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2606.17334","citing_title":"FATE: Pillar Encoding and Frequency-Aware Training for Event-Based Object Detection","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2606.16533","citing_title":"Kairos: A Regret-Aware Native World-Action Model Stack for Physical AI","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2606.12215","citing_title":"MLT-Dedup: Efficient Large-Scale Online Video Deduplication via Multi-Level Representations and Spatial-Temporal Matching","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2606.10303","citing_title":"Isolation-aware Scheduling Framework for DNN-based End-to-End Autonomous Driving System on Tile-based Accelerators","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12608","citing_title":"A Data Efficiency Study of Synthetic Fog for Object Detection Using the Clear2Fog Pipeline","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07965","citing_title":"Zero-Shot Learning in Industrial Scenarios: New Large-Scale Benchmark, Challenges and Baseline","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07953","citing_title":"Unification of Closed-Open Industrial Detection Scenarios: New Large-Scale Benchmarks,Challenges and Baselines","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2606.04992","citing_title":"Multi-Camera AR Guidance System for Surgical Instrument Handling and Assembly: Investigating Workload and Efficiency","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2606.03340","citing_title":"Autonomous Navigation System for Library Service Robot Based on Unitree Go2 Edu","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2606.03694","citing_title":"Face versus Body Tracking for Human-Robot Interaction: An Egocentric Dataset","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21964","citing_title":"Dual-Integrated Low-Latency Single-Lens Infrared Computational Imaging for Object Detection","ref_index":57,"is_internal_anchor":true},{"citing_arxiv_id":"2605.24508","citing_title":"FDDet: Achieving Data-Efficient Food Defect Detection Under Real-World Scenarios","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.24722","citing_title":"Calibrating Probabilistic Object Detectors with Annotator Disagreement","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2606.30476","citing_title":"PS-MOT: Cultivating Instance Awareness from Point Seeds for Multi-Object Tracking","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2606.30215","citing_title":"Efficient RGB-T Object Detection via Sparse Cross-Modality Fusion","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2605.26316","citing_title":"E$^3$C: Video Generation with 3D Environmental Memory and Ego-Exo Human Pose Control","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.29575","citing_title":"Optimizing Latent Representations for Robust Building Damage Assessment Onboard Earth Observation Satellites","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2411.08937","citing_title":"Dual-Head Knowledge Distillation: Enhancing Logits Utilization with an Auxiliary Head","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2412.03603","citing_title":"HunyuanVideo: A Systematic Framework For Large Video Generative Models","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2503.03088","citing_title":"AHCQ-SAM: Toward Accurate and Hardware-Compatible Post-Training Segment Anything Model Quantization","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP","json":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP.json","graph_json":"https://pith.science/api/pith-number/32GMPYZPDA6GC4M2MVJO7WOGEP/graph.json","events_json":"https://pith.science/api/pith-number/32GMPYZPDA6GC4M2MVJO7WOGEP/events.json","paper":"https://pith.science/paper/32GMPYZP"},"agent_actions":{"view_html":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP","download_json":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP.json","view_paper":"https://pith.science/paper/32GMPYZP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.08430&json=true","fetch_graph":"https://pith.science/api/pith-number/32GMPYZPDA6GC4M2MVJO7WOGEP/graph.json","fetch_events":"https://pith.science/api/pith-number/32GMPYZPDA6GC4M2MVJO7WOGEP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP/action/storage_attestation","attest_author":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP/action/author_attestation","sign_citation":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP/action/citation_signature","submit_replication":"https://pith.science/pith/32GMPYZPDA6GC4M2MVJO7WOGEP/action/replication_record"}},"created_at":"2026-07-05T03:03:41.413493+00:00","updated_at":"2026-07-05T03:03:41.413493+00:00"}