Tracked topic
… Two reproduced in-the-wild findings: a silent lost update in ByteDance's deer-flow formalized as a verified L0→L1 fix , and tool-effect reordering in LangGraph's ToolNode on unmodified output removed by an L3 commit-order sequencer . …
Papers arxiv:2605.00503 End-to-End Autoregressive Image Generation with 1D Semantic Tokenizer Published on May 1 Submitted by taesiri on May 4 ByteDance Seed Authors: , , , , , , Abstract End-to-end training of autoregressive image models with joint reconstruction and generation optimization achiev… …
Papers arxiv:2605.31075 Task-Focused Memorization for Multimodal Agents Published on May 29 Submitted by yichen he on Jun 1 ByteDance Seed Authors: , , , , Abstract A reinforcement-learning-based framework called TaskMem is introduced to dynamically determine what information to store in long-term … …
Papers arxiv:2607.05155 EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments Published on Jul 6 Submitted by taesiri on Jul 7 ByteDance Seed Authors: , , , , , , , , , , , , , , , , , , , , , Abstract Analysis of 38,000 hours of real-world agent interactions reveals log-sigmoi… …
Papers arxiv:2606.28436 Dockerless: Environment-Free Program Verifier for Coding Agents Published on Jun 26 Submitted by Yuling on Jul 1 3 Paper of the day ByteDance Authors: , Yuling Shi , , , , , , , , , , , Abstract A Dockerless environment-free agentic patch verifier improves code patch evaluat… …
Papers arxiv:2606.17730 ActWorld: From Explorable to Interactive World Model via Action-Aware Memory Published on Jun 16 Submitted by taesiri on Jun 17 ByteDance Authors: , , , , , , , , , , , , , Abstract ActWorld extends navigation-centric interactive world models to support object interaction th… …
Papers arxiv:2606.29887 SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing Published on Jun 29 Submitted by taesiri on Jun 30 ByteDance Authors: Jiacheng Zhang , , , , , , , Abstract SafePyramid benchmark evaluates guardrail systems' ability to identify safety violations throu… …
Papers arxiv:2605.31604 Representation Forcing for Bottleneck-Free Unified Multimodal Models Published on May 29 Submitted by taesiri on Jun 1 3 Paper of the day ByteDance Seed Authors: Yuqing Wang , , , , , , , , Fuyun Wang , Shuai Wang , , , Abstract Representation Forcing enables unified multimo… …
Papers arxiv:2607.08763 OpenCoF: Learning to Reason Through Video Generation Published on Jul 9 Submitted by taesiri on Jul 10 ByteDance Seed Authors: Xinyan Chen , , , , Abstract OpenCoF framework introduces a reasoning video dataset and model that improve temporal reasoning through diverse superv… …
Papers arxiv:2606.30562 Morphing into Hybrid Attention Models Published on Jun 29 Submitted by Disen Lan on Jul 3 3 Paper of the day ByteDance Seed Authors: Disen Lan , , , , , , , Abstract FlashMorph is an efficient layer selection method that formulates hybrid layer selection as a budget-constrai… …