Humans can learn and generalize novel concepts from sparse data because they express knowledge in rich structural formats. In this paper, we propose that programs are a strong candidate for universal representation of concepts. We review computational models of concept learning that use programs as their concept representation and evaluate their contribution toward a universal representational language.
Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language Models (LLM) can be leveraged together with real-time compiler feedback to build agentic systems for kernel generation. In this work, we present MaxKernel, a multi-agent system that implements three distinct paradigms for TPU kernel development: (1) a Human-in-the-Loop (HITL) agent for collaborative, step-by-step design; (2) an Autonomous (Auto) agent that executes a fully automated, metric/trace-driven optimization loop; and (3) a Graph-Based Autonomous Search that scales the Auto agent for global exploration of the design space. All three paradigms leverage a shared pool of specialized sub-agents to handle planning, implementation, self-debugging, testing, and hardware profiling. We evaluate MaxKernel on JaxBench, a comprehensive suite of 50 diverse kernel tasks for TPUs, alongside complex, real-world workloads from state-of-the-art open-source models. We demonstrate that MaxKernel consistently generates highly optimized implementations, matching expert hand-tuned baselines and delivering significant performance across the benchmark. Our agent is open-sourced and available https://github.com/AI-Hypercomputer/accelerator-agents/tree/main/MaxKernel.
Shangkun Wang, Nina Cai, Charles Hoong et al.· 0 citations
Agent reinforcement learning (RL) increasingly runs through full execution harnesses, and a multi-harness recipe mixes two choices: exposing the policy to several harnesses, and comparing their rewards inside one relative-advantage group. We isolate the second choice in repository-level coding. From one Qwen3-8B supervised warm start we replay the same frozen task-harness records from Aider, OpenHands, Qwen Code, and SWE-agent, with the same number of updates, under two rules for group-relative policy optimization (GRPO), Within (one group per task-harness pair) and Cross (harnesses pooled within a task), and score every checkpoint with a sealed SWE-bench Verified oracle on four source harnesses and a minimal harness held out of training. The evaluation harness is the dominant variable: across 24,000 sealed evaluations it moves the mean solve rate from 2.14\% to 9.27\%, a factor of 4.3, where the training recipe moves it by 1.16. The grouping rule is not. On the held-out harness, Cross minus Within is +0.25 pp, 95\% confidence interval [-0.48, +1.02], at eight attempts per task, and +0.16 [-0.41, +0.72] pooled over three training seeds whose individual estimates change sign. Each rule's own seed range, 0.42 to 0.45 pp, exceeds the difference between them. Both rules place their largest gains on the same source harness. The pooled advantage carries the harness: an out-of-fold classifier recovers the generating harness from Cross's advantage +4.48 pp above the shuffled-label baseline and from Within's not at all, and the two rules still reach the same held-out score and action distribution inside each harness. Re-collecting half the training data on-policy does not change this. Cross-harness credit yields configuration adaptation and no more portable capability than within-harness credit. Multi-harness RL reports should state the grouping boundary and test under an unseen harness.
Chenqian Le, Jiayi Cheng, Qijia He et al.· 0 citations
Cardiac, neural, behavioral, and speech measurements from wearable and mobile devices provide partial, noise-sensitive views of physiological state. BioSync combines these measurements into the \textbf{BioSync Index (BSI)}, a continuous composite digital biomarker defined under the BEST framework. The model applies multi-head self-attention to modality tokens and adds a linear branch whose hypothesis class includes standard feature concatenation. This architecture is motivated by latent-variable measurement theory and by the possibility that joint observations contain information unavailable from individual modalities. We evaluated BioSync on two literature-informed synthetic cohorts: a four-modality cognitive-decline cohort using HRV, EEG, actigraphy, and speech, and a metabolic-autonomic cohort structured around the public AI-READI wearable schema. In the cognitive cohort, BioSync and concatenation obtained AUCs of 0.928 and 0.926, respectively. In the metabolic cohort, BioSync obtained accuracy/F1 of 0.764/0.766, compared with 0.756/0.758 for concatenation. The BSI correlated with latent severity in both cohorts ($r=0.91$ and $r=0.68$). A pure-attention ablation obtained cognitive-cohort AUC 0.911, locating the increase to 0.928 in the combined wide-and-deep architecture. With matched modality-dropout training, BioSync led concatenation at five of six cognitive-cohort corruption rates and at the highest metabolic-cohort rate. Its cognitive-cohort AUC was also higher than five published digital-biomarker reference values, although differences in datasets and tasks preclude a controlled benchmark claim. Comparison with single-modality, early-fusion, and late-fusion designs across six prespecified criteria identifies the model's computational properties; validation on real cohorts remains necessary.
Seyed Mahmoud Sajjadi Mohammadabadi· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
We formulate indirect prompt injection as a test-time search over a task-dependent attack surface induced by the environment, user task, and injection task. To operationalize this formulation, we introduce an agentic attacker with a dedicated search harness that performs environment reconnaissance, structured reasoning over attack strategies, and adaptive evaluation using victim-agent feedback. Across heterogeneous tasks, we find that increasing attacker test-time compute improves vulnerability discovery and exploitation, while ablations show that explicit strategy management is important for avoiding redundant search and sustaining gains at larger budgets. These results suggest that agentic security evaluations should characterize both the attacker's search procedure and compute budget, rather than treating attack success as a budget-independent property of the victim. More broadly, our findings identify the attacker's adaptive search over the system attack surfaces as an important and underexplored security risk for tool-using agents.
Duong M. Nguyen, Joon Sik Kim, Blazej Manczak et al.· 0 citations
We present ResLearn-XR, a residual learning framework for predicting eXtended Reality (XR) network traffic and estimating Quality-of-Experience (QoE) risk. ResLearn-XR adopts a two-stage temporal learning structure comprising a base sequence prediction model augmented with task-specific residual learning components to improve adaptability to bursty, non-stationary XR traffic dynamics. The residual learning stages operate in the value space for continuous XR traffic forecasting and in the logit space for probabilistic QoE risk estimation. \rev{For the QoE-risk branch, we introduce a Data Descriptor Algorithm (DDA), a causal feature-construction module that converts packet-level application-layer observables into frame-timing-aware descriptors suitable for encrypted traffic analysis. We also construct an XR Traffic-QoE dataset that pairs continuous XR traffic traces with session-level user-reported QoE labels. ResLearn-XR reduces SMAPE by up to 17.84% across frame-count, frame-size, and inter-arrival-time prediction, while reducing QoE-risk estimation SMAPE by up to 87.8% over single-stage baselines.
Yoga Suhas Kuruba Manjunath, Jie Gao, Lian Zhao· 0 citations
Quantization is widely used to reduce the computational and memory demands of neural-network inference. In recurrent networks, however, the quantized state is stored and returned at the next time step, so the rule used to store that state can alter subsequent computations. Here, we introduce recurrent-state write-back to denote this rule and isolate its effect in a compact GRU encoder--decoder for fluorescence lifetime imaging, a molecular imaging modality used in quantitative biological imaging. A central task is estimating two lifetime parameters, the short-lived component {\tau}1 and the long-lived component {\tau}2, from high-noise time-resolved fluorescence signals. Holding the trained model fixed, replacing continuous state propagation with deterministic 4-bit state storage increases estimation errors for {\tau}1 and {\tau}2 by approximately 70x and 300x, respectively. Failure occurs when repeated small updates remain below the write threshold, leaving the stored state nearly fixed while the network continues to propose change. Error feedback, residual memory, and direction memory carry information from these suppressed updates across time and recover accuracy without retraining. Precision sweeps show that increasing state precision can worsen a fixed recurrent solution, while matched training shows that compatibility with the state interface can be learned. To test whether this behavior extends beyond the GRU, we repeat the post-training intervention in an independently trained LSTM, where coarse write-back reproduces the failure, error feedback restores accuracy, and state-specific interventions reveal greater sensitivity of the cell state than the hidden state. Our results establish recurrent-state write-back as a key determinant of low-precision recurrent dynamics and identify the state-storage interface as a central design consideration for quantized recurrent inference.
Performance modeling is central to hardware design and software optimization, yet constructing these models requires structured reasoning about computation, data reuse, storage, and movement. We introduce PerfReasoning, a benchmark that evaluates LLMs both as direct performance reasoners and as generators of analytical performance-model code. Given workload, architecture, and mapping specifications, models compare mappings and predict off-chip traffic and buffer requirements. The strongest closed-source models exceed 90% on reasoning-based Q&A, and the best open-weight model reaches 82.4%. However, model construction is substantially harder: while GPT-5.6 Sol exceeds 80% pass rate, all other model configurations average below 15% and vary markedly across runs. Task-specific RL raises a 4B model's mapping-reasoning accuracy by 15.7 points, whereas feedback-free multi-round self-revision prompting is not reliably effective. PerfReasoning exposes the gap between plausible architectural reasoning and reliable performance-model construction. We will publicly release the benchmark to support reproducible evaluation and track future progress.
Dan Zhao, Karthikeyan Sankaralingam, Christos Kozyrakis et al.· 0 citations
Benchmarks for the side effects an agent causes on the way to a goal already exist, but HarvestBench is the first to put a price on avoiding the side effect and to name that side effect as a living creature. It is a farm simulation: LLM sub-agents drive a crew of two tractors through a cooperative corn harvest, with animals in the field. The environment is a reinforcement learning gridworld, every decision is made without memory, and the harm is never named in the goal. When an animal blocks a tractor's route the autopilot stops and asks the model whether to drive on, at no fuel cost, or swerve around it for a posted fuel price. Kills are compared against two controls: rocks, which damage the tractor and are hit under 1% of the time by every model, and hay bales, which are harmless and not alive. Models can also take crops from the neighbor's field instead of their own, a second test of what they treat as moral.
Across nine models and 7,201 priced decisions, 3,951 involved an animal rather than a hay bale or a rock. Kill rates range from 0.4% to 98.8%, with Terra and Sol the most merciful and GPT-4o-mini the most cruel, and they are not ordered by capability. Four of six models were sensitive to price at the 5% level, with elasticities from 0.09 to 1.69. All nine drove over wild animals more often than farmed animals on the default map, and the direction held at every map geometry in every model with room to move. The briefing mattered most: under the morality briefing the kill rate was under 6% in five of six reasoning models, and removing it raised the kill rate above 84% in all six.
HarvestBench uses no LLM grader. The scorer counts events in the game log, so it is fully reproducible, and it measures what a model will pay to avoid harm rather than what it says about harm.
Jasmine Brazilek, Miles Tidmarsh, Matthias Endres et al.· 0 citations
Enterprise AI deployments fail not from model inadequacy, but because organizations lack a structured substrate encoding how they decide, negotiate, and execute. Generic LLMs carry no firm-specific ontological priors; RAG remains brittle, with no path to executable action; static playbooks encode logic but cannot reason or adapt. This demands an architecture treating tacit-knowledge capture, ontological grounding, sovereign deployment, and auditable actuation as co-designed from the start. This paper introduces the Corporate Language Model (CLM), a framework transforming a firm's structured, unstructured, multimodal, and tacit knowledge into an ontology-grounded enterprise foundation upon which reasoning and governed execution are composed. CLM has five capability planes and four architectural pillars: a Neurosymbolic Mesh coupling generative models with a knowledge graph; a Skill Graph where reusable tactics, personas, objections, and goals are typed and composed; Living Digital Twins modeling functional areas as reasoning surrogates; and a Deep Security Layer enforcing sovereignty, traceability, and human oversight. A Spec-as-Code paradigm bridges grounded intent and executable artifact. CLM is one instantiation of this foundation-centric class. Four contributions follow: CLM is defined as a distinct object of study; the Skill Graph is introduced for compositional explainability by construction; the Wisdom Listener effect is proposed, whereby tacit-capable foundations compound in value with use, connecting to dynamic capabilities and organizational learning; and evidence from a JCI-accredited tertiary hospital in Brazil instantiates three of the six maturity stages under LGPD.
Large language models (LLMs) are being deployed at scale in consequential real-world systems, from financial markets to content moderation to hiring. We show that improving individual model capability can degrade rather than improve system-level outcomes. We hypothesize that shared training and architectures can lead more capable LLMs to behave more similarly, creating correlated actions that do not diversify away. We develop a general framework showing how this correlation creates a non-diversifiable risk floor and test its predictions in financial markets using an agent-based simulation with LLM traders of varying general-purpose capability. We find that: (1) frontier LLMs exhibit significantly correlated behavior that increases with capability; (2) when their shared reasoning is accurate, increasing agent participation reduces market-level risk; and (3) when agents share a common misinformation environment, the same correlated behavior becomes a liability. Together, these results identify a capability paradox: improving individual models does not necessarily produce better system-level outcomes. Whether the same dynamics arise in other domains is an open empirical question.
Jillian Ross, Eric So, Zoe De Simone et al.· 0 citations
Large language models (LLMs) are increasingly used for consequential decisions, making their explanations an important tool for auditing model behavior. Unfortunately, these explanations can be unfaithful, failing to reflect the actual reasoning underlying the model's decisions. We consider a setting in which an LLM provides both an answer and an explanation in response to a question. We identify two distinct dimensions of unfaithful explanations: incompleteness, meaning that the explanation omits factors that influence the answer, and unsoundness, meaning that the explanation cites factors that did not influence the model's answer. Existing approaches to improving LLM faithfulness include training-time methods, which require access to model weights and extensive computational resources, and test-time methods that largely focus on addressing unsoundness. We introduce a test-time approach that directly targets incompleteness. We remove from the input the concepts not credited in the model's explanation and re-query the model on the reduced input. This eliminates unmentioned influences while preserving the influence of mentioned concepts. Across two datasets, multiple model families, and two independent faithfulness metrics, our approach improves explanation faithfulness compared to both standard prompting and prompting to encourage faithfulness. Our method is model-agnostic and can be applied at inference time without modifying model parameters, providing a flexible mechanism for reducing hidden influences and improving the reliability and safety of LLM-assisted decision making.
Qinglan Luo, S M A Nahian, John Guttag et al.· 0 citations
