TechniqueRLHF / Alignment8 recent entries7 Oct 2026CroissantMiner: Automated Extraction and Validation of Croissant Metadata for ML DatasetsarXiv:2610.07132v1 Announce Type: cross Abstract: Croissant has emerged as a standard for machine-readable dataset metadata, yet populating its fields remains labor-intensive and requires careful read→8 Oct 2026TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence ReasoningarXiv:2610.09360v1 Announce Type: cross Abstract: Real-world documents distribute evidence across text, tables, figures, and captions within complex page layouts. Answering complex questions over such
TechniqueRAG8 recent entries1 Oct 2026Drift Inspector: Exploring and Measuring Scientific Drift with Atomic Contribution ClaimsarXiv:2609.39710v1 Announce Type: new Abstract: Scientific abstracts mix contributions with background, motivation, and meta-language, so tools that read them as-is cannot separate what a field produc→4 Oct 2026Synthesis: Arxiv-Cs-ClAuto-generated synthesis of 10865 entries about arxiv-cs-cl→4 Oct 2026Wiki Lint Report — 2026-10-04Automated lint: 19 errors, 64915 warnings, 3 info→5 Oct 2026Are you Synthesizing or Recalling? Evaluating LLMs on Algorithmic Code RetrievalarXiv:2610.02438v1 Announce Type: cross Abstract: Large language models (LLMs) have demonstrated strong performance in code generation, where success depends on both recalling relevant algorithmic kno→8 Oct 2026TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence ReasoningarXiv:2610.09360v1 Announce Type: cross Abstract: Real-world documents distribute evidence across text, tables, figures, and captions within complex page layouts. Answering complex questions over such→9 Oct 2026DaCe-DT: Data-Centric Offline Multi-Task Reinforcement Learning via Adaptive Prompts and Trajectory Correction for Heterogeneous TasksarXiv:2610.11085v1 Announce Type: new Abstract: Offline multi-task reinforcement learning (Offline MTRL) heavily depends on the quality and distribution of pre-collected data. However, existing method→13 Oct 2026Machine Learning Tutorial/Workshop [D]This post is to let people know about a workshop/tutorial on Machine Learning that will run on the (blue) math discord server: discord.gg/maths. The place is nice, people are cool and helpful, and as →14 Oct 2026Retrievable Gradients: Continual Post-Training Without Cumulative Weight DriftarXiv:2606.15734v2 Announce Type: replace-cross Abstract: Continual post-training enables models to absorb emerging knowledge after deployment, but repeatedly updating shared parameters can accumulate
TechniqueAgents8 recent entries13 Oct 2026Brilliant paper on agentic continual learning from Meta and MIT. (bookmark it) Agent memory harnesses usually only write forward. A lesson g…Brilliant paper on agentic continual learning from Meta and MIT. (bookmark it) Agent memory harnesses usually only write forward. A lesson goes into the bank and is never checked against whether it he→14 Oct 2026Trust Before Autonomy: Rethinking Agentic Remote Sensing for Crisis ResponsearXiv:2610.18789v1 Announce Type: new Abstract: Agentic remote sensing is evaluated on how many questions it can answer, whereas crisis response depends on whether its answers can be relied on and how→14 Oct 2026Square’s new kiosk lets you talk to AI for help with your orderSquare is launching new hardware: a big kiosk for ordering things that has a touchscreen and lets you talk to an AI. It's called the Square Kiosk. Square says the Kiosk, which has a large pill shape, →14 Oct 2026Interesting new research from Meta Superintelligence Labs It discusses how reward hacking affects coordination in multi-agent coding systems…Interesting new research from Meta Superintelligence Labs It discusses how reward hacking affects coordination in multi-agent coding systems. Coding agents on a team share code, memory, and plans, so →14 Oct 2026EAU: Environment-Agent-User Co-evolution for Training User-Centric AgentsarXiv:2610.16523v1 Announce Type: cross Abstract: Training LLM agents to interact with users over multiple turns is challenging because the learning signal depends not only on the agent, but also on t→14 Oct 2026Bridging Risk Approximation Gaps in Model Predictive Task Sampling via In-Context ModelingarXiv:2610.14913v1 Announce Type: cross Abstract: Reinforcement learning (RL) is essential for adaptive decision-making, e.g., robust Meta-RL and domain randomization (DR), and efficient post-training→14 Oct 2026AI Discovery in Combinatorics: From Problems to Solutions and Back to Bigger ProblemsarXiv:2610.16555v1 Announce Type: cross Abstract: We introduce qtBench, a benchmark of 29 open problems in algebraic combinatorics. These problems ask for a simple combinatorial algorithm that explain→15 Oct 2026Giving my Dot a dedicated email address turned out to be surprisingly painless & solved a few problems around how to relate to a personal ag…Giving my Dot a dedicated email address turned out to be surprisingly painless & solved a few problems around how to relate to a personal agent. I don't want it using my email or drafting messages on
TechniqueFine-tuning8 recent entries5 Oct 2026Unmasking Propaganda: A Comparative Analysis of Masked and Causal Language ModelsarXiv:2610.03077v1 Announce Type: new Abstract: Propaganda detection is an essential task in natural language processing (NLP), particularly in the context of manipulative political communications. Ho→5 Oct 2026Keep the Effect, Drop the Actor: Programmable Effect-to-Execution World-Action ModelsarXiv:2610.02398v1 Announce Type: new Abstract: A robot demonstration records two things in the same frames: what happened to the objects, and how one particular arm made it happen. We condition on th→6 Oct 2026Self-Reflection Fine-Tuning: Enhancing Agent Security against Prompt Injection Attacks from Failure ExperiencearXiv:2610.04269v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly deployed in tool-augmented environments, but their reliance on external inputs makes them highly vuln→7 Oct 2026A stock GLM-Edge-1.5B-Chat on a 4GB Galaxy A04e completed a real Amazon cart taskYesterday TechCrunch published a piece about a growing problem for AI agents: websites are starting to block them. Amazon blocking Meta's Muse is the obvious example. At almost exactly the same time, →8 Oct 2026SkillSandbox: Skill Verification via Dynamic Scenario SynthesisarXiv:2610.10088v1 Announce Type: cross Abstract: Self-evolving agents distill task-solving experience into skills for future reuse, but these skills can encode incorrect procedures or non-transferabl→8 Oct 2026MIRROR: From Imitation to Internalization in LLM PersonalizationarXiv:2610.09795v1 Announce Type: new Abstract: The demand for personalized LLMs is shifting from style imitation toward content quality. We investigate whether self-distillation can bridge this gap i→8 Oct 2026Big new paper nobody is talking about ‼️ 🌎 Robot world models now have scaling laws. Meta trained an 8B-parameter JEPA on 15,000 hours of r…Big new paper nobody is talking about ‼️ 🌎 Robot world models now have scaling laws. Meta trained an 8B-parameter JEPA on 15,000 hours of robot video to prove it. RoboJEPA (FAIR at Meta + Mila) imagin→9 Oct 2026Easy to anticipate, hard to compute: boundary dependence finds the computed outputs that entropy patching missesarXiv:2610.11790v1 Announce Type: new Abstract: Byte-level language models such as the Byte Latent Transformer (BLT) group bytes into patches and run their large global model once per patch. BLT start
TechniqueMultimodal8 recent entries9 Oct 2026Can LLMs Fix It Without Code? Toward Automated Verification of No-Code Bug FixesarXiv:2610.11963v1 Announce Type: cross Abstract: A no-code fix resolves an invalid bug report by directing the user to change a setting, update to a version where the problem is already fixed, or adj→12 Oct 2026Evo-World: A Lightweight Vision-Language-Action Model with Future Motion Prediction via Optical-Flow SupervisionarXiv:2610.12936v1 Announce Type: new Abstract: Vision-language-action (VLA) models provide a unified framework for visual understanding, language grounding, and robot control, yet anticipating how th→12 Oct 2026CORE-Seg: Reasoning-Driven Segmentation for Complex Lesions via Reinforcement LearningarXiv:2603.05911v2 Announce Type: replace-cross Abstract: Medical image segmentation is undergoing a paradigm shift from conventional visual pattern matching to cognitive reasoning analysis. Although →14 Oct 2026Variance Fingerprints: Component-Wise Identification of Content Factors by Augmentation DesignarXiv:2610.14260v1 Announce Type: new Abstract: Augmentation-based self-supervised learning can recover the content shared across views without separating its underlying factors into individual coordi→14 Oct 2026Trust Before Autonomy: Rethinking Agentic Remote Sensing for Crisis ResponsearXiv:2610.18789v1 Announce Type: new Abstract: Agentic remote sensing is evaluated on how many questions it can answer, whereas crisis response depends on whether its answers can be relied on and how→14 Oct 2026Towards Modality-Agnostic Seizure Classification: A Unified Deep Learning Framework for Heterogeneous EEG and Intracranial EEG SignalsarXiv:2610.16596v1 Announce Type: new Abstract: Seizure prediction and detection, essential for clinical treatment of epilepsy, remain the crucial task of this neurological disease. EEG and iEEG are w→14 Oct 2026Meta-Video: Efficient Standardized Evaluation for Video UnderstandingarXiv:2610.18535v1 Announce Type: new Abstract: Video understanding benchmarks are the primary yardstick for measuring progress in video-language modeling, and a growing body of work relies on them to→14 Oct 2026dMRI-F: A Foundation Model for Generalizable Diffusion MRI Analysis from Sparse AcquisitionsarXiv:2610.17968v1 Announce Type: new Abstract: Diffusion MRI (dMRI) probes water diffusion within the brain, providing unique information about tissue microstructure and white matter connectivity. Hi
TechniqueSafety8 recent entries12 Oct 2026Agreement Without Alignment: Label-Free Rubric Clarification Increases LLM-Judge Consensus but Not Agreement with HumansarXiv:2610.13303v1 Announce Type: cross Abstract: LLM judges are increasingly deployed as panels, and disagreement among panel members is often read as a symptom of an unclear rubric. We test a label-→13 Oct 2026Inside Action Gateway in Managed AgentsA coding agent finishes a change and wants to open a GitHub issue with a link to its results. The agent can describe the issue, but creating it belongs to a different system. It needs a tool that can →14 Oct 2026Who Improves the Robot? Toward Physical Recursive Self-ImprovementarXiv:2610.18842v1 Announce Type: cross Abstract: Robots are increasingly autonomous in task execution, yet their improvement remains largely human-driven. This study investigates Physical Recursive S→14 Oct 2026Variance Fingerprints: Component-Wise Identification of Content Factors by Augmentation DesignarXiv:2610.14260v1 Announce Type: new Abstract: Augmentation-based self-supervised learning can recover the content shared across views without separating its underlying factors into individual coordi→14 Oct 2026OneDemo: Unified Policy Learning and Data Generation from a Single DemonstrationarXiv:2610.18188v1 Announce Type: new Abstract: Learning robot policies from demonstrations is often limited by the cost of collecting diverse, high-quality expert data. While an appealing alternative→14 Oct 2026FENAS: A Few-Shot Meta-Ensembling Framework for Efficient Neural Architecture Search with Synergized Zero-Cost ProxiesarXiv:2610.15307v1 Announce Type: new Abstract: Efficient neural architecture search (NAS) requires accurate performance estimation under limited evaluation budgets. Existing zero-cost proxies (ZCPs) →14 Oct 2026EAU: Environment-Agent-User Co-evolution for Training User-Centric AgentsarXiv:2610.16523v1 Announce Type: cross Abstract: Training LLM agents to interact with users over multiple turns is challenging because the learning signal depends not only on the agent, but also on t→14 Oct 2026Bridging Risk Approximation Gaps in Model Predictive Task Sampling via In-Context ModelingarXiv:2610.14913v1 Announce Type: cross Abstract: Reinforcement learning (RL) is essential for adaptive decision-making, e.g., robust Meta-RL and domain randomization (DR), and efficient post-training