11725 papers
FlowCorrect is a modular, interactive imitation learning framework that enables real-time, sample-efficient adaptation of generative flow-matching robotic policies through sparse human pose corrections, significantly improving success rates on previously failed tasks without retraining the underlying model.
This paper introduces AOT-SFT, a large-scale adversarial dataset, and AOT, a self-play framework that co-evolves an image-editing attacker with a defender MLLM to dynamically generate training data, thereby significantly enhancing the model's perceptual robustness and reducing hallucinations in complex visual scenes.
This paper introduces GR4AD, a production-oriented generative recommendation system for large-scale advertising that integrates novel tokenization, a lazy autoregressive decoder, and value-aware reinforcement learning to achieve significant revenue gains and high-throughput real-time serving in Kuaishou's 400-million-user ecosystem.
This paper introduces AMA-Bench, a novel benchmark designed to evaluate long-horizon memory in agentic applications using real-world and synthetic machine-generated trajectories, and proposes AMA-Agent, a causality-driven memory system that significantly outperforms existing baselines by addressing the limitations of current similarity-based retrieval methods.
This paper introduces the CTFIDU+ algorithm to establish the completeness of identifying counterfactual queries from physically realizable Layer 3 data, defines the fundamental limits of exact causal inference in this setting, and derives novel analytic bounds for non-identifiable quantities that are empirically shown to be tighter with counterfactual data.
This paper addresses the evaluation gap in music generation by introducing CMI-RewardBench, a comprehensive ecosystem comprising large-scale datasets, a unified benchmark, and efficient reward models to evaluate and improve music generation under complex compositional multimodal instructions.
This study introduces causal circuit tracing to reveal that distinct single-cell foundation models (Geneformer and scGPT) share conserved computational architectures characterized by inhibitory dominance and biological coherence, with cross-model consensus identifying disease-associated domains that are validated by CRISPRi as reflecting co-expression rather than causal encoding.
This paper introduces a qualitative content analysis-based framework for annotating economic narratives as directed acyclic graphs and demonstrates through a factorial experiment that locally-constrained representations and overlap-based metrics significantly improve inter-annotator agreement by reducing human label variation.
This paper proposes a sample-efficient tensor factorization model that combines cheap automated ratings with a small set of human labels to enable fine-grained, accurate, and confidence-interval-backed evaluation of generative models at the prompt level.
This paper establishes Federated Inference as a distinct collaborative paradigm for privacy-preserving model serving by formalizing its core requirements, analyzing structural trade-offs, and identifying system-level challenges that differentiate it from federated learning and classical ensemble methods.
This paper demonstrates an exponential gap between structured and unstructured pruning within the Strong Lottery Ticket Hypothesis, proving that while unstructured weight pruning requires only logarithmic overparameterization to approximate a target neuron, structured neuron pruning necessitates linear overparameterization.
This paper presents a unified study that systematically investigates the interactions between the temperature parameter and various training components in knowledge distillation, offering practical guidance for selecting optimal temperature values to improve student performance.
This paper proposes a new paradigm for causal discovery that leverages crowdsourcing, expert elicitation, and LLM-based simulation to aggregate fragmented knowledge from multiple agents into a comprehensive global causal structure unattainable by any single entity.
This paper introduces Probe-Select, a plug-in module that predicts final image quality from early denoising activations to enable efficient early termination of unpromising seeds, thereby reducing sampling costs by over 60% while improving the quality of retained images in text-to-image generation.
This paper introduces a reinforcement learning framework for Markov Decision Processes with exogenous dynamics that leverages the independence of certain state components from agent actions to achieve information-theoretically optimal regret bounds and significantly improved sample efficiency compared to standard methods.
AriadneMem is a structured memory system for long-horizon LLM agents that employs a decoupled two-phase pipeline of entropy-aware filtering, conflict-aware coarsening, and algorithmic bridge discovery to significantly improve multi-hop reasoning accuracy and efficiency while drastically reducing context usage and runtime.
This paper presents a hybrid LLM architecture and evaluation framework (DG-EVAL) that combines supervised fine-tuning on expert-curated agricultural facts with a safety-aware stitching layer to deliver accurate, culturally appropriate, and cost-effective conversational advisory for smallholder farmers in India.
This paper proposes TTSR, a test-time self-evolving framework where a single language model alternates between student and teacher roles to analyze reasoning failures and generate targeted variant questions, thereby enabling stable and continual improvement in reasoning performance on challenging benchmarks.
This paper addresses the NP-hard challenge of optimal offline semantic caching for LLM embeddings by proposing polynomial-time heuristics and novel online policies that leverage recency, frequency, and locality to improve response speed, reduce costs, and enhance semantic accuracy.
This paper proposes a dual-stream transformer architecture that unifies language and structured data processing by encoding knowledge graphs and hypergraphs into a separate key-value repository, where a journey-based role transport mechanism enables the language model to attend over structured instances while maintaining a clear, inspectable separation between linguistic context and structured knowledge.