Knowledge catalogue
arXiv:2605.27752v1 Announce Type: new Abstract: LLM confidence calibration is often evaluated by comparing two signals: token-probability scores and verbalized confidence. These signals are sometimes
arXiv:2603.00349v2 Announce Type: replace Abstract: Many complex tasks require extended effort, diverse capabilities, or coordinated actions beyond what a single agent can provide. However, simply add
arXiv:2605.28144v1 Announce Type: new Abstract: LLMs have shown remarkable proficiency in general language understanding and reasoning. However, they consistently underperform in spatial reasoning tha
arXiv:2605.27559v1 Announce Type: cross Abstract: Multi-stage LLM pipelines that perform multi-agent debate, intrinsic self-correction, or retrieval-augmented verification exhibit puzzling aggregate b
arXiv:2605.27784v1 Announce Type: new Abstract: LLM agents are governed by long-lived natural-language prompt policies, but individually reasonable standing rules can interact in uninspected ways. We
Every LLM API call depends on the inference engine underneath it. Tokenization, scheduling, prefill, decode, KV cache, batching, and streaming determine whether the experience is fast, scalable, and p
arXiv:2605.28302v1 Announce Type: cross Abstract: Modern large language model (LLM) inference has progressively disaggregated to keep pace with growing model sizes and tight TTFT and TPOT service-leve
arXiv:2605.27375v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly acting as autonomous agents, but their continuous interaction with the environment can lead to in-context
arXiv:2605.28524v1 Announce Type: new Abstract: In recent years, Large Language Models (LLMs) have shown great capability in processing graph tasks such as fraud detection. However, most existing meth
arXiv:2509.23019v5 Announce Type: replace-cross Abstract: Watermarking offers a promising solution for detecting LLM-generated content, yet its robustness under realistic query-free (black-box) evasio
arXiv:2605.28168v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated promising capability in generating reward functions for deep reinforcement learning (DRL)-based building
arXiv:2511.14584v3 Announce Type: replace-cross Abstract: We present ReflexGrad, a dual-process architecture for within-episode failure recovery in LLM agents without demonstrations. When agents commi
arXiv:2605.28791v1 Announce Type: cross Abstract: On-policy self-distillation (SD) improves LLM reasoning by using teacher-side privileged information (PI) to turn sparse verifier outcomes into dense
arXiv:2605.27401v1 Announce Type: cross Abstract: There is a growing interest in utilizing synthetic populations for a diverse range of applications. At the same time, we are witnessing a tremendous g
arXiv:2511.11896v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have recently shown strong potential in vulnerability detection (VD). However, accurately detecting vulnerabiliti
arXiv:2605.28224v1 Announce Type: new Abstract: Multi-trajectory inference for tool-use LLM agents - generating multiple reasoning attempts and selecting among them - benefits from transferring knowle
arXiv:2605.27913v1 Announce Type: new Abstract: Node classification on graphs often requires labeled nodes, yet obtaining labels at graph scale is expensive. When node attributes contain semantic cont
arXiv:2605.26596v1 Announce Type: new Abstract: The token-level extractive compressors widely used for general LM context are structurally inappropriate for LLM agents: across 17 (env, backbone, metho
arXiv:2602.17443v2 Announce Type: replace Abstract: Multi-turn LLM evaluation is typically reported as a single win-rate scalar, conflating distinct capabilities. We introduce AIDG (Adversarial Inform
arXiv:2605.26492v1 Announce Type: cross Abstract: LLM-generated stories are a popular use case, but they show very low variability. We sample 20,000 total stories from four current models using five p
arXiv:2605.26121v1 Announce Type: cross Abstract: LLM pre-training efficacy increasingly depends on data composition rather than sheer volume. Yet, optimal mixing is hindered by categorization flaws:
arXiv:2411.02355v4 Announce Type: replace-cross Abstract: Quantization is a powerful tool for accelerating large language model (LLM) inference, but the accuracy-performance trade-offs across differen
arXiv:2605.27204v1 Announce Type: new Abstract: Scientific paper evaluation often involves not only assessing a manuscript itself, but also relating it to contemporaneous research and prior literature
arXiv:2605.26731v1 Announce Type: new Abstract: A prevalent assumption in LLM agent deployment holds that more structured harnesses universally improve reliability, and that higher-capability models n
arXiv:2605.26154v1 Announce Type: cross Abstract: LLM-driven agents are capable of selecting external tools to complete users' tasks. However, attackers could compromise such process, steering agents
arXiv:2605.26878v1 Announce Type: new Abstract: Multi-stakeholder tasks require one output to satisfy users with conflicting preferences. Holistic LLM judges conflate utility estimation and utility ag
arXiv:2603.14864v2 Announce Type: replace Abstract: In e-commerce, LLM agents show promise for shopping tasks such as recommendations, budget management, and bundle deals, where accurately capturing u
The MiniMax M2 series was one of the most widely used open-weight LLM series earlier this year. Now, we got a technical report with some interesting tidbits. I summarized some of them below: 1. Full a
arXiv:2605.26433v1 Announce Type: new Abstract: Large language model (LLM) summarization systems may pass compact vector representations of private inputs to downstream retrieval, monitoring, audit, o
arXiv:2605.23916v1 Announce Type: cross Abstract: LLM tool registries function as unregulated advertising platforms: providers write free-text descriptions that agents use for selection, yet no measur
arXiv:2605.25338v1 Announce Type: cross Abstract: Large language model (LLM) agents frequently fail on multi-step tasks involving reasoning, tool use, and environment interaction. While such failures
arXiv:2605.24786v1 Announce Type: cross Abstract: Long-horizon LLM inference turns the key--value (KV) cache into the dominant GPU memory consumer and makes per-token attention increasingly expensive.
arXiv:2605.24137v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used to generate summaries of software bug reports, including sections such as Steps-to-Reproduce (S2R),
arXiv:2605.24144v1 Announce Type: cross Abstract: Large Language Models (LLMs) have achieved impressive performance across diverse domains but remain inefficient during the autoregressive decoding pha
arXiv:2605.25389v1 Announce Type: cross Abstract: While Large Language Model-based Multi-Agent Systems (LLM-MAS) demonstrate remarkable capabilities in solving complex tasks by orchestrating specializ
arXiv:2605.24613v1 Announce Type: cross Abstract: Post-hoc repair of LLM mathematical reasoning introduces an asymmetric risk: fixing an incorrect reasoning trace is useful, but replacing a trace that
arXiv:2605.25203v1 Announce Type: cross Abstract: We apply the influence-adaptive Walsh geometry of a companion theory paper (arXiv:2605.01637) to extreme low-bit weight-only LLM quantization. The rec
arXiv:2605.24043v1 Announce Type: cross Abstract: Scientific discovery is a closed-loop process in which hypotheses guide data acquisition and observations refine the hypothesis space. Yet most approa
arXiv:2605.24490v1 Announce Type: new Abstract: Multi-agent LLM decision systems for portfolio management still lack a principled way to assign credit across specialist agents, remain vulnerable to co
arXiv:2605.24614v1 Announce Type: cross Abstract: Large language model (LLM) unlearning has emerged as a crucial post-hoc mechanism for privacy protection and AI safety, yet auditing whether target kn
arXiv:2605.24941v1 Announce Type: cross Abstract: Modern LLM agents combine long-term memory for personalization with tool-calling interfaces for taking actions in the world -- a combination underpinn
arXiv:2605.24048v1 Announce Type: cross Abstract: Multi-AI collaboration, such as ensembling or debating large language models (LLMs), is a promising paradigm for aggregating information and boosting
arXiv:2505.23803v2 Announce Type: replace-cross Abstract: Phishing email detection faces significant challenges due to evolving adversarial tactics and heterogeneous attack patterns. Traditional appro
arXiv:2605.24319v1 Announce Type: new Abstract: As large language models become a default source of guidance on personal, moral, and existential questions, it matters whether they draw on the religiou
arXiv:2605.10977v2 Announce Type: replace-cross Abstract: Watermarking for large language models (LLMs) is a promising approach for detecting LLM-generated text and enabling responsible deployment. Ho
arXiv:2605.25424v1 Announce Type: cross Abstract: Existing LLM routing frameworks treat queries as independent events, neglecting the sequential nature of real-world user sessions constrained by globa
arXiv:2605.25850v1 Announce Type: cross Abstract: This paper investigates large language model (LLM) abstention learning, specifically using ternary reward, which incentivize truthfulness in large lan
arXiv:2605.24052v1 Announce Type: cross Abstract: To better serve users' demands in mobile applications (e.g., navigation), mobile crowdsourcing platforms can iteratively align large language model (L
arXiv:2509.23573v4 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly used to help security analysts manage the surge of cyber threats, automating tasks from vulnerab
arXiv:2603.06610v2 Announce Type: replace Abstract: Large language model (LLM) post-training enhances latent skills, unlocks value alignment, improves performance, and enables domain adaptation. Unfor
arXiv:2605.23362v1 Announce Type: new Abstract: Evaluating large language models increasingly relies on LLM-as-a-judge protocols, but such evaluations remain costly: different judges have different pr
arXiv:2605.23175v1 Announce Type: cross Abstract: Proprietary large language models (LLMs) face risks of intellectual property (IP) violation, as adversaries can replicate an LLM by collecting input-o
arXiv:2605.23414v1 Announce Type: new Abstract: LLM-based multi-agent systems can fail even when planned actions are executed correctly because agents may misjudge their knowledge when evaluating plan
LLM companies are likely to be like airline companies: small margins, intense competition, high expenses. @GaryMarcus Airlines taught us something important: unlimited demand does not guarantee attrac
not using LLM’s works for me Prolonged AI use may make it harder to think critically and creatively, recent research suggests. But there are ways to keep the brain fit https://www.economist.com/scienc
arXiv:2605.22566v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents demonstrate strong reasoning and execution capabilities on complex tasks when guided by structured instructions,
oh even better, since you can test the runtime without an LLM, you can post this into Claude: “look up http://activegraph.ai, install and run a small experiment I would like leveraging this, and expla
arXiv:2605.21851v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards has become the standard recipe for improving LLM reasoning, but the dominant algorithm GRPO assigns a sin
arXiv:2605.22263v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) is an emerging LLM post-training paradigm in which the model serves as its own teacher: conditioned on privileged inf
arXiv:2605.20734v1 Announce Type: cross Abstract: A large language model (LLM) agent that sends messages can leak data inside them. Destination allowlists and content scanners do not police whether an