Natural-Language Agent Harnesses
arXiv:2603.25723v2 Announce Type: replace-cross Abstract: Agent performance is strongly shaped by the surrounding harness: the external execution system around a model that organizes a task run. Yet t
Knowledge catalogue
arXiv:2603.25723v2 Announce Type: replace-cross Abstract: Agent performance is strongly shaped by the surrounding harness: the external execution system around a model that organizes a task run. Yet t
arXiv:2501.06933v3 Announce Type: replace Abstract: Nonlinear conservation laws govern a broad class of important physical systems in science and industry and are central to scientific machine learnin
arXiv:2605.16418v1 Announce Type: cross Abstract: EEG-based visual decoding aims to establish a mapping between neural signals and visual semantics. However, it remains constrained by the dual challen
arXiv:2605.18396v1 Announce Type: new Abstract: Video generation models produce visually compelling results but systematically violate physical commonsense -- on VideoPhy-2, the best model achieves on
arXiv:2603.03308v2 Announce Type: replace-cross Abstract: How does the conversational past of large language models (LLMs) influence their future performance? Recent work suggests that LLMs are affect
arXiv:2605.17862v1 Announce Type: cross Abstract: Scaling on-policy distillation (OPD) for large language models (LLMs) confronts a fundamental tension: asynchronous execution is necessary for system
arXiv:2601.21531v2 Announce Type: replace-cross Abstract: Visual token compression is widely used to accelerate large vision-language models (LVLMs) by pruning or merging visual tokens, yet its advers
arXiv:2508.16438v4 Announce Type: replace-cross Abstract: Recent advances in large language models (LLMs) and dense retrievers have driven significant progress in retrieval-augmented generation (RAG).
arXiv:2307.12405v2 Announce Type: replace Abstract: We propose a machine learning approach to the optimal control of multiclass fluid queueing networks (MFQNETs) that provides explicit and insightful
arXiv:2605.16395v1 Announce Type: cross Abstract: We present OrbiSim, a novel robotic simulation paradigm that redefines world models as a fully differentiable physics engine for embodied intelligence
arXiv:2605.17877v1 Announce Type: new Abstract: A significant hurdle for current LLMs is the execution of complex, multi-stage tasks. Group Relative Policy Optimization (GRPO) has been emerging as a l
arXiv:2605.18191v1 Announce Type: new Abstract: Current reinforcement learning(RL) methods are broadly applicable and powerful in verifiable settings where scalar rewards can be provided. However, in
arXiv:2605.18080v1 Announce Type: cross Abstract: We present a novel parameterized 4-qubit Eisert-Wilkens-Lewenstein (EWL) quantum game circuit for recommender systems in quadruple helix innovation ec
arXiv:2605.18038v1 Announce Type: new Abstract: Salmon re-identification in commercial net-pens is challenging due to large populations, which impose strict accuracy requirements and make large-scale
arXiv:2605.16452v1 Announce Type: cross Abstract: Accurate peak detection across diverse cardiac physiological signals, including the Electrocardiogram (ECG), Photoplethysmogram (PPG), Ballistocardiog
arXiv:2605.17671v1 Announce Type: cross Abstract: Non-contrastive self-supervised learning (SSL) is an effective framework for predictive representation learning, but popular (and in practice effectiv
arXiv:2505.00409v3 Announce Type: replace-cross Abstract: Automatic anonymization is increasingly used to enable ethical sharing of clinical speech, yet its perceptual and clinical consequences remain
arXiv:2605.18303v1 Announce Type: cross Abstract: World models built on recurrent state space architectures enable efficient latent imagination, yet remain physically unstructured, producing dynamics
arXiv:2605.16325v1 Announce Type: cross Abstract: Phase-transition phenomena in deep learning (grokking, emergent capabilities, and ontological reorganization under context shift) have been studied th
arXiv:2605.18606v1 Announce Type: new Abstract: Neural operators approximate PDE solution maps, but they need not respect the symmetries of the governing equation. In out-of-distribution (OOD) regimes
arXiv:2605.16860v1 Announce Type: cross Abstract: Accurate long-horizon glucose forecasting is critical for automated insulin delivery systems, which help people with type 1 diabetes (T1D) manage thei
arXiv:2605.16351v1 Announce Type: cross Abstract: Scientific foundation models are expected to reuse representations under changes in dataset, acquisition protocol, and deployment domain, yet many seq
arXiv:2605.18598v1 Announce Type: new Abstract: We address the fundamental question of why deep neural networks generalize by establishing a pointwise generalization theory for fully connected network
arXiv:2605.16479v1 Announce Type: cross Abstract: Job seekers often initiate search with short, underspecified queries. At LinkedIn, over 80% of job-related queries contain three or fewer keywords, ma
Politicians who boost AI do not represent the views of their constituents. New polling in the UK shows: - Most people think AI will destroy more jobs than it creates (57% v 17%) - 7 in 10 are worried
arXiv:2605.17347v1 Announce Type: cross Abstract: When a neural network estimates someone's age from a photograph, does it process biometric data? The answer depends on whether identity-discriminative
arXiv:2605.18352v1 Announce Type: new Abstract: Presupposition projection in conditionals is central to theories of meaning and pragmatics, yet it remains largely unevaluated in large language models.
arXiv:2605.16793v1 Announce Type: new Abstract: Time series forecasting under non-stationarity faces a fundamental tension between capturing stable representations and adapting to distribution shifts.
arXiv:2605.17382v1 Announce Type: new Abstract: The rapid progress of generative artificial intelligence has exposed fundamental limitations in existing evaluation methodologies, particularly for open
arXiv:2605.16429v1 Announce Type: cross Abstract: We introduce extbf{QuantFPFlow}, a reinforcement learning framework that integrates quantum amplitude estimation into the Fokker--Planck~(FP) formulat
arXiv:2605.18570v1 Announce Type: new Abstract: Cross-domain knowledge alignment is essential for integrating heterogeneous medical systems, yet existing approaches typically treat entity alignment as
arXiv:2605.18130v1 Announce Type: new Abstract: Medical image segmentation is more clinically valuable when it supports diagnosis rather than merely producing lesion masks. However, diagnostically rel
arXiv:2605.18591v1 Announce Type: cross Abstract: Natural policy gradients improve optimization by accounting for the geometry of distribution space, but their practical use is limited by the cost of
arXiv:2605.16999v1 Announce Type: new Abstract: Reinforcement learning post-training has substantially improved the reasoning accuracy of vision-language models, yet the resulting policies remain poor
arXiv:2605.18359v1 Announce Type: new Abstract: Large multimodal models (LMMs) inherit the self-attention mechanism of pretrained language backbones, yet standard attention can exhibit suboptimal allo
arXiv:2601.22664v4 Announce Type: replace Abstract: Reinforcement Learning from Human Feedback (RLHF) is a pivotal technique for aligning large language models (LLMs) with human preferences, yet it is
arXiv:2605.17308v1 Announce Type: new Abstract: Electrocardiogram (ECG) diagnosis in clinical practice relies on structured reasoning over multiple hierarchical aspects, including cardiac rhythm, cond
arXiv:2605.16460v1 Announce Type: new Abstract: Referring expression counting (REC) is an intention-driven task that requires context-aware visual reasoning. While recent vision-language models incorp
arXiv:2605.16302v1 Announce Type: cross Abstract: Reinforcement learning for multi-step reasoning with large language models (LLMs) often relies on sparse terminal rewards, leading to poor credit assi
arXiv:2605.17036v1 Announce Type: new Abstract: This paper studies autonomous generative AI agents in multi-echelon supply chains using the MIT Beer Game. We identify four inference-time levers that s
arXiv:2502.07027v2 Announce Type: replace-cross Abstract: Molecular Relational Learning (MRL) is widely applied in natural sciences to predict relationships between molecular pairs by extracting struc
arXiv:2605.18599v1 Announce Type: new Abstract: Transformer-based models have advanced feedforward novel view synthesis (NVS). Current architectures such as GS-LRM and LVSM mix semantic information (e
arXiv:2605.17548v1 Announce Type: cross Abstract: Code review has evolved for decades, from informal peer checking to today's pull request (PR) workflows, yet it remains a largely manual, uneven, and
arXiv:2605.16331v1 Announce Type: cross Abstract: Protein language models are increasingly used to guide experimental and clinical decisions, yet it is often unclear whether a confident prediction ref
arXiv:2605.16651v1 Announce Type: new Abstract: Explanation mechanisms are increasingly used to support transparency and trust in vision-language models (VLMs), particularly in settings where model de
arXiv:2605.18004v1 Announce Type: new Abstract: Randomized linear algebra (RLA) algorithms are a modern class of numerical linear algebra techniques that play an essential role in scientific computing
Running the same CNA search on base models (before instruction tuning) yields a structurally similar set of neurons, but ablating them produces almost no behavioral change. We read this as evidence th
arXiv:2605.18579v1 Announce Type: new Abstract: Pre-training on text-attributed graphs (TAGs) is central to building transferable graph foundation models, where LLM-as-Aligner methods align graph and
arXiv:2605.16871v1 Announce Type: new Abstract: Explainable robots require not only successful task execution but also the ability to expose internal decision-making process in a user-friendly manner.
arXiv:2605.17648v1 Announce Type: new Abstract: Generative recommendation treats next-item prediction as autoregressive item-identifier generation. Specifically, items are encoded as semantic identifi
arXiv:2605.17271v1 Announce Type: cross Abstract: Bi-causal optimal transport (OT) is a natural framework for comparing and coupling stochastic processes under nonanticipative information constraints,
arXiv:2605.17084v1 Announce Type: cross Abstract: In language models, what a representation encodes is determined by the geometry of its representation space: distances, not activations, carry meaning
arXiv:2605.18299v1 Announce Type: new Abstract: Search-augmented reasoning agents interleave internal reasoning with calls to an external retriever, and their performance relies on the quality of each
arXiv:2605.18162v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have made striking progress, yet their spatial reasoning remains fragile: models that answer an original input correctly
arXiv:2605.18039v1 Announce Type: new Abstract: Learning dense correspondences across deformable 3D shapes remains a long-standing challenge due to structural variability, non-isometric deformation, a
arXiv:2605.17999v1 Announce Type: new Abstract: This paper proposes a Shared Backbone Proximal Policy Optimization (Shared Backbone PPO) algorithm. By sharing the base module between the Actor and Cri
arXiv:2605.16973v1 Announce Type: new Abstract: Domain generalization aims to enhance model robustness against unseen domains with embedding distribution shifts. While large-scale vision-language mode
This post appears to be a community prompt-sharing discussion where Swyx invites people to share their '/goal' prompts (likely system prompts or directives designed to optimize AI behavior toward spec
arXiv:2605.17850v1 Announce Type: cross Abstract: iffusion-based generative models increasingly rely on inference-time guidance, adding a drift term or reweighting mixture of experts, to improve sampl
arXiv:2605.17341v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have achieved remarkable success, yet their reliance on massive datasets and unintended memorization of training data ra