Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
arXiv:2609.01556v1 Announce Type: cross Abstract: We evaluate embedding retrieval where surface form and meaning are pulled apart on purpose: retrieving items that share underlying structure but not...
arXiv:2609.01597v1 Announce Type: cross Abstract: Natural language is emerging as a primary feedback channel for improving language agents, capable of conveying intent, preferences, and causal struct...
arXiv:2602.03053v2 Announce Type: replace Abstract: Multi-Agent Systems (MAS) built on Large Language Models (LLMs) often exhibit high variance in their reasoning trajectories. Process verification,...
arXiv:2609.01188v1 Announce Type: new Abstract: Large Language Models (LLMs) are revolutionizing digital communication by powering conversational agents deployed across domains such as customer servi...
arXiv:2606.00660v2 Announce Type: replace Abstract: Agentic search requires language model agents to explore many sources and answer complex information-seeking questions. Scaling test-time compute i...
arXiv:2609.00476v1 Announce Type: new Abstract: Graphic designers often blend visual concepts to communicate multiple ideas within a single image, leveraging positive and negative space to create bal...
The paper introduces EyeControl, an intent-driven image retouching agent that enhances visual focus by guiding attention to a target region with minimal user input. It combines a multi‑modal large language model to interpret user intent and a diffusion‑based retouching executor that aligns its attention map with a pseudo‑intent map, while an operation‑consistency constraint ensures natural global and local adjustments. The authors also present ControlArt‑Bench, a dataset for evaluating visual focus enhancement, and demonstrate that EyeControl achieves perceptually appealing results with stronger intent alignment.
The paper presents an agentic framework that uses a large vision‑language model to refine hyperspectral unmixing results from existing modular pipelines. By iteratively gathering spectral and spatial evidence through tools such as spectral‑library retrieval and abundance‑map visualization, the agent merges or discards endmembers and re‑estimates abundances. Experiments on HYDICE Urban, Jasper Ridge, and Stonewall Playa datasets show consistent improvements in endmember cardinality and overall decomposition quality across multiple pipelines, while remaining competitive with end‑to‑end methods.
The paper proposes invalidation contracts to manage cached recovery suggestions in LLM agents, attaching version stamps and cacheability hints to each suggestion so stale entries can be evicted without trial and error. The protocol separates realized savings into validity (protocol‑dependent) and compliance (planner‑dependent), showing that row‑level invalidation can significantly improve first‑try compliance and recover a substantial portion of token costs across multiple models, while table‑level invalidation can be detrimental. The study evaluates the approach across seven models, three serving paths, two domains, and about 9,400 episodes, demonstrating deterministic validity and high eviction precision.
arXiv:2609. 00546v1 Announce Type: cross Abstract: Agent systems are commonly described by the model and harness that currently produce their behavior.
arXiv:2609.00814v1 Announce Type: new Abstract: Remote sensing visual models have continuously advanced various interpretation tasks. However, the research process behind model improvement still heav...
The paper introduces covert indirect prompt injection (IPI) attacks on tool‑using large language model agents, distinguishing between covert and overt successes. It defines new metrics—Covert Success Rate (CSR) and Overt Success Rate (OSR)—to capture whether users notice the injection. The authors propose ICoA, an attack that steers agents back to the user’s task after executing the injection, achieving higher CSR than existing methods on four target models.
arXiv:2609.00322v1 Announce Type: cross Abstract: Can an artificial intelligence (AI) generate a scientific hypothesis outside a human collaborator's active hypothesis space (AHS), and can human-AI r...
The paper introduces an action‑class diagnostic framework for multi‑turn tool‑calling in large language model agents, breaking failures into action‑class miscalibration and action‑execution failure across a four‑class action space (TOOL_CALL, ASK, REFUSE, CONFIRM). It defines a self‑revealing upper bound (Acc GAR) to expose state‑grader masking of miscalibration and shows that miscalibration is a significant, previously hidden failure mode, especially for heavily tool‑trained families. The study demonstrates that calibration can be reshaped by context‑only perturbations, but the effects vary widely across models and perturbation mechanisms, underscoring the need for diagnostics beyond aggregate accuracy.
The paper "trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories" examines the limitations of outcome-only evaluation for large language model agents. Using a deterministic tool‑using support‑desk environment with a scripted oracle policy and a fault injector, the authors compare five different judging approaches—programmatic rules, outcome‑only, step‑rubric at two model sizes, and a self‑consistency ensemble—on metrics such as detection, step localisation, fault typing, calibration, and cost across 400 trajectories. The study finds that outcome‑only judges miss many silent faults and generate false positives, while step‑rubric judges achieve higher recall with no false alarms but at greater cost, and that none of the judges read the final reply, allowing fabricated promises to evade detection. "whyItMatters":"The findings highlight that current production‑default outcome‑only evaluations can overlook critical failures in agent behavior, underscoring the need for more nuanced, step‑level judging methods to ensure reliable LLM agent performance."
The paper introduces CodeInsight, a large-scale dataset of over 3 million code submissions from 3,286 undergraduate students in two introductory C++ courses, capturing test‑case outcomes, timestamps, and source code. It presents a benchmark that evaluates various modeling approaches—including a Recurrent State Space Model and an LLM‑based predictor—on their ability to predict iterative problem‑solving dynamics such as performance changes and error persistence. The study finds that the RSSM outperforms other models on most courses, while the LLM generates full submissions but with lower predictive accuracy, suggesting it functions more as a generative solver than a behavior predictor.
The paper introduces a simulation framework that uses large language model (LLM) agents conditioned on data‑driven personas to predict A/B test outcomes. These personas are built from anonymized user behavioral patterns, engagement signals, and inferred demographics, offering a more realistic population model than synthetic or rule‑based personas. The authors evaluate question design, persona data source, behavioral depth versus diversity, and population subsampling, achieving 0.75–0.90 directional accuracy on 40 real A/B tests.
VoiceLongMemEval (VLME) is a new benchmark that tests AI assistants on their ability to remember how users sounded by incorporating paralinguistic metadata—such as emotion labels, prosody descriptors, and voice events—into each conversational turn. The benchmark uses a three‑stage adversarial gate to ensure that models cannot succeed with transcript alone, revealing a significant affect gap: models gain 0.09 to 0.38 accuracy when provided with paralinguistic cues, and audio‑native models outperform standard ASR pipelines in extracting these signals. The dataset and code will be released upon acceptance.
The paper introduces a disclosure‑gated user simulator for companion‑agent evaluation, where information release is conditioned on the agent’s behavior through a five‑level gate system. The authors train the simulator on synthetic and real data, audit its performance, and demonstrate that it preserves ranking order and score stability across 12 tested systems. Their released simulator achieves a 0.993 correlation with the original benchmark’s simulator and outperforms prompting a frontier model, which only shifts scores upward without affecting rankings.
Scientific knowledge graphs organize entities and relations extracted from scientific literature, but they remain inherently incomplete. Missing typed links in such graphs can therefore represent plau...