Category
20,463 papers
Agent harness evolution enables recursive self-improvement (RSI) without updating model weights. We examine whether higher benchmark scores yield reliable, efficient, and transferable improvement. We evaluate four methods across harnesses, general and coding tasks. Repeated evaluations, controlled agent expansion, comp...
The domain of a fine-tune trained on a previously unseen dataset is recoverable well above chance from gray-box activations and black-box outputs, and activations and outputs were more informative than the weight-delta statistics.
The theoretical analysis clarifies the role of basis consistency and supports the proposed consistency-aware aggregation mechanism, FedAbA, which is proposed as a basis-consistent federated LoRA aggregation framework for heterogeneous ranks.
FlashOffloading is presented, a host-DRAM-centric inference architecture that fully offloads the KV cache in both the prefill and decode phases, optimizing KV offloading latency across the entire pipeline and delivering throughput gains under a time-per-output-token (TPOT) service-level objective (SLO).
Reach audiences
Advertise in front of researchers, engineers, and readers.
CST-Bench is introduced, a benchmark for counterfactual spatial reasoning and CT-OPSD is proposed, which uses CST-Bench annotations to guide a teacher model in learning counterfactual state transitions and causal reasoning chains and provides an initial demonstration of how its annotations can support model improvement...
Goal-Branching is proposed, where blocking agents are given temporary goals, each tried as an additional branch of LaCAM's search, and random and corridor-aware rules for choosing these temporary goals are examined and it is shown that completeness is preserved for any such rule.
A rolling-horizon distributionally robust model that updates predictions of the remaining closure duration as observations of disruption resolution become available is developed, which minimizes the worst-case expected total completion delay over an ambiguity set of scenario probabilities.
DelegateEval is proposed, a platform-level evaluation framework for agent systems that may combine large language models, reusable skills, protocols such as MCP, and executable tool environments and assess whether a platform enables people with different backgrounds to turn their expertise into ongoing services that de...
This work formalizes LRM safety objectives as requiring non-malicious reasoning and answers, safety consistency, and explicit intent identification, and introduces a reasoning-preserving gradient projection mechanism that analytically resolves conflicts between safety and reasoning gradients, effectively eliminating th...
This work introduces FINCOSMOS, an open-world evaluation paradigm for agents doing financial work, and shows how an agent with persistent memory learned all ten tasks with FINCOSMOS feedback.
Results show that how information is represented to an agent shapes both its cost and its success, making the environment worth optimizing in addition to the parameters and the harness, making the environment worth optimizing in addition to the parameters and the harness.
When a tropical cyclone (TC) threatens several ports, preparedness planning depends on anticipating which ports are likely to be affected and where disruptions may last longest. Limited historical observations at individual ports and uncertainty in landfall location make these relative impacts difficult to estimate. We...
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
