Knowledge catalogue
arXiv:2605.20693v1 Announce Type: new Abstract: Interpretable text representations should expose coordinates that are not only predictive, but also meaningful enough for independent auditors to apply.
arXiv:2605.20244v1 Announce Type: cross Abstract: We present Lean Refactor, a plug-and-play retrieval-augmented agentic framework for multi-objective, controllable, and version-robust refactoring of L
arXiv:2605.19321v1 Announce Type: cross Abstract: Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe responses, motivating pre-model and post-model guards.
arXiv:2605.19194v1 Announce Type: new Abstract: The Mixture-of-Agents (MoA) framework has shown promise in improving large language model (LLM) performance by aggregating outputs from multiple agents.
arXiv:2403.07183v4 Announce Type: replace-cross Abstract: We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a lar
arXiv:2605.19633v1 Announce Type: cross Abstract: Can a single LLM-based optimization system match specialized tools across fundamentally different domains? We show that when optimization problems are
arXiv:2510.18821v3 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) has become the mainstream technique for training LLM agents. However, RLVR highly depends on w
Self-improving AI is a big deal! As a first step, I've been exploring how much of the post-training can be automated. Here is a first post on how I am using @FireworksAI_HQ Agent to automate LLM fine-
arXiv:2605.18792v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) improves large language models (LLMs) by incorporating external evidence, but it also introduces knowledge confli
arXiv:2601.22569v2 Announce Type: replace-cross Abstract: Large language model (LLM) based agents are increasingly used to automate financial transactions, yet their reliance on contextual reasoning e
arXiv:2605.17437v1 Announce Type: cross Abstract: Context. Metamorphic Testing addresses the test-oracle problem in scientific computing, but classical Mutation Score operates on syntactic AST mutatio
arXiv:2510.04309v3 Announce Type: replace Abstract: Controlling the behaviors of large language models (LLM) is fundamental to their safety alignment and reliable deployment. However, existing steerin
arXiv:2510.03879v3 Announce Type: replace-cross Abstract: Translating C to memory-safe languages, like Rust, prevents critical memory safety vulnerabilities that are prevalent in legacy C software. Ev
arXiv:2605.16462v1 Announce Type: cross Abstract: Detecting unauthorized knowledge distillation from a deployed LLM API is hard because the defender controls neither the attacker's training pipeline n
arXiv:2502.18632v4 Announce Type: replace Abstract: Knowledge components (KCs) mapped to problems help model student learning, tracking their mastery levels on fine-grained skills thereby facilitating
arXiv:2605.16704v1 Announce Type: new Abstract: Improving LLM performance on downstream tasks sometimes requires leveraging auxiliary datasets during post-training. In practice, however, developers fa
arXiv:2604.01658v2 Announce Type: replace Abstract: Large language model (LLM)-based evolution is a promising approach for open-ended discovery, where progress requires sustained search and knowledge
arXiv:2605.17439v1 Announce Type: cross Abstract: Evaluating LLM-generated interactive software requires execution in addition to static analysis. The key difficulty is that correctness is a graph-lev
arXiv:2605.18421v1 Announce Type: cross Abstract: Recent benchmarks for Large Language Model (LLM) agents mainly evaluate reasoning, planning, and execution. However, memory is also essential for agen
arXiv:2605.16615v1 Announce Type: new Abstract: In many applications, human and LLM evaluators use assessments of relevant criteria to create an overall evaluation for an item or individual. For examp
arXiv:2605.17757v1 Announce Type: cross Abstract: INT2 KV-cache quantization is attractive for long-context LLM serving, but it remains difficult to make both accurate and deployable. Simple rotations
arXiv:2602.03352v2 Announce Type: replace Abstract: Reinforcement learning (RL) has shown strong promise for LLM-based machine translation, with recent methods such as GRPO demonstrating notable gains
arXiv:2605.17072v1 Announce Type: new Abstract: Existing LLM-driven knowledge graph (KG) construction methods predominantly employ stateless batch processing pipelines, exhibiting structural deficienc
arXiv:2605.18579v1 Announce Type: new Abstract: Pre-training on text-attributed graphs (TAGs) is central to building transferable graph foundation models, where LLM-as-Aligner methods align graph and
arXiv:2605.17790v1 Announce Type: new Abstract: LLM-based equation discovery offers a promising route to recovering symbolic laws from data, but many systems still rely on generation-centered loops th
arXiv:2605.16699v1 Announce Type: new Abstract: Capped-usage SaaS products -- LLM subscriptions such as Claude Code and ChatGPT, cloud platforms such as Vercel and Cloudflare Workers, corporate benefi
arXiv:2505.19241v2 Announce Type: replace-cross Abstract: The recent success in using human preferences to align large language models (LLMs) has significantly improved their performance in various do
arXiv:2605.15759v1 Announce Type: new Abstract: Large language model (LLM) agents require long-term memory to leverage information from past interactions. However, existing memory systems often face a
arXiv:2605.15223v1 Announce Type: cross Abstract: This paper presents an LLM-empowered workflow for RISC-V supply chain analysis, integrating Vision-Language Models (VLMs) and Model-Driven Engineering
arXiv:2605.15226v1 Announce Type: cross Abstract: We ask whether agentic AI systems built for software engineering transfer to realistic hardware engineering. Existing hardware LLM benchmarks isolate
it’s funny how people here just make stuff up. @GaryMarcus Take away LLM & every single AI application today goes back to the stone age, driverless cars will immediately break down, all the apps will
arXiv:2605.15508v1 Announce Type: cross Abstract: The quadratic complexity of attention imposes severe memory and computational bottlenecks on Large Language Model (LLM) inference. This challenge is p
arXiv:2605.14236v1 Announce Type: cross Abstract: Pairwise Ranking Prompting (PRP) elicits pairwise preference judgments from an LLM, which are then aggregated into a ranking, usually via classical so
arXiv:2605.14401v1 Announce Type: cross Abstract: Memory-augmented LLM agents have advanced personalized recommendation, yet existing approaches universally adopt flat memory representations that conf
arXiv:2605.13940v1 Announce Type: cross Abstract: Third-party skills are becoming the package ecosystem for LLM agents. They package natural-language instructions, helper scripts, templates, documents
arXiv:2605.15187v1 Announce Type: new Abstract: A bottleneck in learning to understand articulated 3D objects is the lack of large and diverse datasets. In this paper, we propose to leverage large lan
arXiv:2605.14344v1 Announce Type: new Abstract: Generative modeling has emerged as a promising approach for crystal structure discovery. However, existing LLM-based generative models struggle with low
arXiv:2605.14445v1 Announce Type: new Abstract: Many real-world coding challenges are open-ended and admit no known optimal solution. Yet, recent progress in LLM coding has focused on well-defined tas
arXiv:2605.14005v1 Announce Type: new Abstract: Speculative decoding has become a widely adopted technique for accelerating large language model (LLM) inference by drafting multiple candidate tokens a
arXiv:2605.14089v1 Announce Type: new Abstract: In recent years, a variety of powerful LLM-based agentic systems have been applied to automate complex tasks through task orchestration. However, existi
arXiv:2605.14700v1 Announce Type: new Abstract: Generating robot simulation environments remains a major bottleneck in simulation-based robot learning. Constructing a training-ready MuJoCo scene typic
arXiv:2605.14290v1 Announce Type: cross Abstract: ReAct has become the default architecture across LLM agents, and many existing web agents follow this paradigm. We argue that it is the wrong default
arXiv:2605.13044v1 Announce Type: cross Abstract: LLM-powered agents can silently delete documents, leak credentials, or transfer funds on a routine user request, not because the agent was attacked, b
arXiv:2605.12784v1 Announce Type: new Abstract: Advances in large language models (LLMs) have recently opened new and promising avenues for small-molecule drug discovery. Yet existing LLM-based approa
arXiv:2605.11442v1 Announce Type: cross Abstract: Large Language Model (LLM) agents have emerged as key intermediaries, orchestrating complex interactions between human users and a wide range of digit
arXiv:2605.12477v1 Announce Type: cross Abstract: LLM-based agents increasingly operate in persistent environments where they must store, update, and reason over information across many sessions. Whil
arXiv:2605.12492v1 Announce Type: new Abstract: We introduce Pion, a spectrum-preserving optimizer for large language model (LLM) training based on orthogonal equivalence transformation. Unlike additi
arXiv:2605.11534v1 Announce Type: new Abstract: When an LLM-based embodied agent fails at a household task, the culprit could be misidentified objects, forgotten sub-goals, or poor action sequencing -
arXiv:2605.06375v1 Announce Type: cross Abstract: Large language model (LLM) alignment via reinforcement learning from human preferences (RLHF) suffers from unstable policy updates, ambiguous gradient
arXiv:2605.10380v1 Announce Type: new Abstract: LLM-based agents deliver state-of-the-art performance across tasks but incur high end-to-end latency on edge devices. We introduce Agent-X, a software-o
arXiv:2605.08715v1 Announce Type: cross Abstract: LLM-based multi-agent systems are increasingly deployed on long-horizon tasks, but a single decisive error is often accepted by downstream agents and
arXiv:2605.08761v1 Announce Type: cross Abstract: Large language model (LLM) agents are increasingly expected to operate in enterprise environments, where work is distributed across specialized roles,
arXiv:2605.09584v1 Announce Type: cross Abstract: Inpatient clinical reasoning is a sequential decision under partial observability: the clinician sees the admission so far and must choose the next ac
arXiv:2602.02350v2 Announce Type: replace Abstract: Multi-Agent Discussion (MAD) has garnered increasing attention very recently, where multiple LLM instances collaboratively solve problems via struct
arXiv:2605.10488v1 Announce Type: cross Abstract: Agent-compiled knowledge bases provide persistent external knowledge for large language model (LLM) agents in open-ended, knowledge-intensive downstre
arXiv:2605.09959v1 Announce Type: cross Abstract: Self-evolving LLMs excel in verifiable domains but struggle in open-ended tasks, where reliance on proxy LLM judges introduces capability bottlenecks
arXiv:2605.09942v1 Announce Type: new Abstract: Memory retrieval in agentic large language model (LLM) systems is often treated as a static lookup problem, relying on flat vector search or fixed binar
arXiv:2510.17671v2 Announce Type: replace-cross Abstract: Many real-world optimization problems are guided by complex, subjective preferences that are difficult to express as explicit closed-form obje
arXiv:2510.22170v2 Announce Type: replace Abstract: Persona conditioning is widely used to steer large language model (LLM) behavior, but it is unclear whether it induces stable behavioral structure o
arXiv:2605.08221v1 Announce Type: cross Abstract: This paper presents NoisyCoconut, a novel inference-time method that enhances large language model (LLM) reliability by manipulating internal represen