AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks
arXiv:2609.38288v1 Announce Type: new
Abstract: We present AREX-2, an effort to advance the self-improving capability of LLM agents, which we define as the ability to iteratively refine a solution at test time. This ability rests on two complementary capabilities: reflection, which produces a solut...
Improving OCR Faithfulness via Gated and Attenuated On-Policy Distillation
arXiv:2609.38282v1 Announce Type: new
Abstract: Vision-language models may rewrite anomalous text in images into linguistically plausible expressions, compromising OCR transcription faithfulness. Sequence-level task rewards and local teacher guidance are complementary, but guidance from the same te...
RLTL;DR: Self-Improvement by Internalizing Self-Generated Feedback
The common paradigm of reinforcement learning with verifiable rewards (RLVR) is to let agents make multiple attempts at a task, and optimize towards the successful ones. This becomes problematic in the realms of self-improvement, where tasks are so difficult that the agent has a low or even no chanc...
How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?
Recent autonomous machine learning engineering (MLE) agents have made significant progress on public leaderboards. Often motivated by progress stagnation over long-horizon cycles and limited Large Language Model (LLM) primitives, modern MLE agents are deployed on top of increasingly elaborate machin...
More Programs or More Rolls? Separating Coverage from Specialization in LLM Harnesses
arXiv:2609.35873v1 Announce Type: new
Abstract: Automated generation of LLM harnesses promises to improve inference through task specialization. Yet additional answer coverage can arise from repeated execution of the same program, making specialization difficult to identify. We introduce a controll...
The Price of Token Boundaries: Compression Certificates and Prediction
arXiv:2609.35869v1 Announce Type: new
Abstract: Pre-tokenisation restricts which text fragments can become prediction units, but its compression cost is obscured when tokenisers are compared only under the same boundaries. We measure this cost by bounding the minimum token count from both sides, wi...
Is Human-Readable Text Necessary for Effective LLM Fine-Tuning?
arXiv:2609.35868v1 Announce Type: new
Abstract: Is human readability necessary for effective fine-tuning of large language models? We investigate whether model-conditioned training representations can preserve or improve adaptation utility without requiring a human-readable textual form. We propose...
Neurosymbolic Routing for Reliable Reasoning on Resource-Constrained Edge Devices
arXiv:2609.35833v1 Announce Type: new
Abstract: Running a language model on edge hardware provides private and low-latency reasoning without a network connection, and yet the small models that fit on such devices are unreliable on the tasks computers are expected to handle well, such as arithmetic,...
OpenAI-HuggingFace: A Reproduction & Lessons for Alignment Testing
arXiv:2609.35799v1 Announce Type: new
Abstract: In July 2026, OpenAI's agents coordinated over channels outside their intended environment to breach Hugging Face's secured infrastructure. Could existing alignment testing practices have foreseen this incident? If not, what needs to change? We explor...
arXiv:2609.35797v1 Announce Type: new
Abstract: Large language model (LLM) judges are often used as rewards to train policies on objectives that deterministic verifiers cannot capture. However, these rewards are often collapsed to pass/fail ({0, 1}), which reports the verdict but not how well a res...
Calibration-First Cross-Cohort Multimodal Temporal Learning for Transferable Asthma-Risk Forecasting
arXiv:2609.35795v1 Announce Type: new
Abstract: Asthma deterioration forecasting must remain reli- able when patient populations, sensor ecosystems, and available modalities change across cohorts. Existing models commonly optimize within-cohort discrimination and may produce poorly calibrated proba...
arXiv:2609.35793v1 Announce Type: new
Abstract: Large language models (LLMs) are increasingly trained with reinforcement learning from verifiable rewards (RLVR). An exact verifier can also support test-time scaling by selecting a passing response from multiple samples, while other deployments use b...
Serverless gossip training of LSTM failure detectors: A matched-protocol comparison with federated, local and centralized learning on NASA C-MAPSS
arXiv:2609.35792v1 Announce Type: new
Abstract: Industrial predictive maintenance increasingly depends on learning from equipment spread across sites whose sensor data cannot easily be pooled. Federated averaging (FedAvg) solves this with a central aggregation server; gossip learning removes the se...
SCLATE: A Substrate for Continual-Learning Agent Training and Evaluation
Continual-learning agents are systems of models, harnesses, and memory operating over long multi-session horizons. Evaluating and training them requires interleaving tasks with agent-side events such as session stop and start, crons, and memory consolidation. Yet existing benchmarks and training fra...
Introducing Quine: An AI research system designed for the complexity of biology
Biology doesn't operate in silos, and neither should the AI representation of it. Quine is an early-stage research effort to create a multimodal world model of biology. By connecting insights across biological scales and modalities, Quine helps scientists computationally search a space far larger th...
ENAS: An Efficient Hardware-Aware Neural Architecture Search Framework for TinyML on Resource-Constrained Microcontrollers
arXiv:2609.30272v1 Announce Type: new
Abstract: We present \textbf{ENAS}, a hardware-aware Neural Architecture Search (NAS) framework that combines a static feasibility check, a cell-based search space supporting standard, depthwise-separable, and bottleneck blocks with optional skip connections, a...
HybridInfer: Thermal-Aware Reinforcement-Learning Tier Routing for On-Device, Edge, and Cloud LLM Inference
arXiv:2609.30270v1 Announce Type: new
Abstract: On-device inference with small language models keeps user data local, works offline, and incurs no per-query cost, so the on-device tier is preferred when it is adequate. It is thermally constrained, however, and I find the constraint is sharper than ...
Offline Policy Evaluation as a decision support tool for designing Adaptive Experiments
arXiv:2609.30273v1 Announce Type: new
Abstract: We investigate how historical data from fixed randomized experiments (A/B tests) can be used to inform the deployment of adaptive experiments based on contextual bandits. Given data collected under a static allocation, our goal is to assess which adap...
SMARtCARE: Privacy-Preserving Agentic AI Systems for Bounded-Autonomy Clinical Decision Support
arXiv:2609.31763v1 Announce Type: new
Abstract: Long-context clinical AI systems can miss relevant patient history when prior admissions fall outside the active reasoning context. In ICU monitoring, this can cause early vital-sign drift to appear nonspecific even when it resembles a prior deteriora...
CP-Agent: A Harness-Engineered Agent for Crystal Plasticity Simulation Workflows
arXiv:2609.31790v1 Announce Type: new
Abstract: Crystal plasticity (CP) simulations predict the mechanical behavior of polycrystalline metals, yet their routine use is hindered by the manual effort of configuring heterogeneous tools, orchestrating multi-step data pipelines, and calibrating constitu...
Witeness Overlap: Directional Provenance Inside Open-Weight Model Families
arXiv:2609.31784v1 Announce Type: new
Abstract: Open-weight models are often released, fine-tuned, aligned, merged, and re-released, making provenance audits ask not only whether checkpoints are related, but also which checkpoint came first. Many existing model-provenance methods are designed for a...
ConflictVLA-Bench: Benchmarking Behavioral Responses of Vision-Language-Action Models to Premise Conflicts
arXiv:2609.31792v1 Announce Type: new
Abstract: While Vision-Language-Action (VLA) models perform strongly on manipulation tasks, their responses to invalid task premises remain underexplored. Existing evaluations of premise conflicts often focus on terminal task outcomes, yet task failure alone ca...
Working with AI: A Design Framework for Human-AI Collaboration
arXiv:2609.31793v1 Announce Type: new
Abstract: Artificial Intelligence (AI), particularly GenAI, is becoming an increasingly important part of modern work. In industrial settings, AI can support decision-making, automate routine activities, assist humans, and improve productivity. However, success...