1. Memory-based self-improving agents--those that learn from an online stream of tasks and improve over time by maintaining a textual memory bank--have shown great promise in recent literature. However, the reliability aspects of these method…

  2. We introduce PTXBench, a benchmark for evaluating and adapting large language models (LLMs) to use architecture-specific PTX for GPU kernel optimization. PTXBench measures functional correctness, whether selected target instructions execut…

  3. Recent advances in Large Language Models(LLMs) and agents have substantially improved the ability of AI systems to execute complex tasks. Yet existing benchmarks largely rely on researcher-selected tasks, leaving uncertain whether such pro…

  4. Large language model-based agentic frameworks primarily optimize capability: whether an agent can reason, retrieve information, call tools, delegate work, and complete a goal. Enterprise execution requires a stronger property.

  5. As AI datacenters retire functional GPUs, vast quantities of still capable accelerators enter secondary markets. This paper investigates whether these retired GPUs can find a productive afterlife to form a DumpsterCluster that can serve mo…

  6. Agentic applications are shifting AI serving from isolated model inference to long-running workloads in which LLMs coordinate tools, environments, and persistent state. However, the system behavior of these workloads---where latency, cost,…

  7. LLM-based agents can act on behalf of a user to access cloud services, call tools, or invoke agents. At session start, the agent's permissions are set but remain static, and each request is evaluated independently, without considering prio…

  8. When a user question is underspecified, a capable model should recognize that its context is insufficient, identify the missing information, ask for it, and respond only once that information determines a unique answer. We formalize multi-…

  9. Frontier LLM agents increasingly transact on behalf of separate principals, often using natural language rather than structured APIs. Much of the safety literature studies misaligned LLM behavior through adversarial-elicitation evaluations…

  10. Large Language Model (LLM) serving has become a critical cloud workload, and realistic traces are essential for motivating and benchmarking serving systems. However, existing LLM serving workload studies remain limited in scale and scope.

  11. As large language models scale, their training-token budgets must also increase to maintain an appropriate tokens-per-parameter ratio (\(\mathrm{TPP}\)). However, high-quality domain data is much harder to scale than general web data.

  12. Patients increasingly ask large language model (LLM) assistants which doctor to see, making these systems AI infomediaries: algorithms that intermediate one person's choice among other people and thereby decide, silently and at scale, whic…

  13. Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human des…

  14. There are 56,804 public agent skills today, and teams write many more privately. The dominant delivery model is installation: once installed, a skill's description remains in the system prompt, competing for fewer than 100 reliable trigger…

  15. Professional communication is increasingly mediated by LLMs - but do these models serve all users equally? We show that when prompts contain linguistic features more commonly used by women (hedges, tag questions, collective reference), the…

  16. Self-improving LLM agents convert successful trajectories into persistent cross-task state. An unsafe success can thereby become reusable policy after its triggering input disappears.

  17. This paper reports a single, fully instrumented case study of a large-scale architectural refactoring by an AI coding agent under a specification-first protocol, with no human review of the generated code and no pre-existing oracle to vali…

  18. Systems that generate GPU kernels with language models report high correctness rates. Those rates come from a single loose test: run the kernel on a few random inputs at one fixed shape and accept it if the output is close to a reference.

  19. GPU kernel agents and GPU programming languages have advanced separately, leaving expert kernels difficult to reproduce. Agents usually treat the compiler as a fixed black box and receive only errors, correctness outcomes, and timing, whil…

  20. The dominant paradigm treats AI safety as a property to be instilled during model training via RLHF, DPO, or Constitutional AI. We argue this is structurally insufficient for autonomous agents that execute code, mutate files, send messages…

  21. LLM-agent services repeatedly execute small deterministic transitions between model and tool calls: route an outcome, update state, and emit the next effect. We ask when this control path exposes enough concurrent work for GPU execution, a…

  22. A benchmark score comes from a single phrasing of each problem. That single phrasing is treated as if it stood for the whole space of ways the same problem could be asked, but it does not.

  23. Agentic coding READMEs like this http URL grow without bound in real repositories, stopping only when the repository retires or someone rewrites the file wholesale. We trace this to imperfect recall: appending an instruction is always chea…

  24. An agent skill is a folder containing a this http URL file with instructions for a language-model agent, optionally accompanied by scripts and reference files. The agent loads the skill when it judges that a task matches the skill descript…

  25. Over the past several years, LLM-powered chatbots and agents have become widely used as a tool for academic writing. LLM-assisted writing can be valuable by removing language barriers but at the same time causes concerns about misconduct a…

  26. As AI coding agents take on increasingly complex, long-horizon software engineering tasks, existing benchmarks are rapidly saturating and their evaluation quality has come under serious scrutiny: a recent audit found that nearly 60% of uns…

  27. LLM routers promise efficiency by matching each request to the cheapest adequate model, and are increasingly applied per step inside multi-step agents. Yet agentic routers are evaluated like single-turn routers: by replaying logged traject…

  28. Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth. Dense attention gives each token broad horizontal access to the past, but the vertical feedback channel between d…

  29. We present Ouroboros, a self-developing agent harness whose tools, prompts, context assembly, and core implementation improve through reviewed commits that become the runtime for later work. Core evolution proceeds in two modes.

  30. AI agents are becoming shared infrastructure, yet durable memory is commonly assembled from separate retrieval, governance, and operational components. We present SuperLocalMemory 4.0, a governed, local-first memory operating system for AI…