AWARE augments generative next-POI recommendation with LLM agents that produce user-anchored narratives capturing events, culture, and trends, delivering up to 12.4% relative gains on three real datasets.
hub
Advances in neural information processing systems , volume=
27 Pith papers cite this work. Polarity classification is still indexing.
27
Pith papers citing it
hub tools
representative citing papers
OLIVIA treats LLM agent action selection as a contextual linear bandit over frozen hidden states and applies UCB exploration to adapt online, yielding consistent gains over static ReAct and prompt-based baselines on four benchmarks.
LLMs copy biased analyst ratings in investment decisions but a new detection method encourages independent reasoning and can improve stock return predictions beyond human levels.
CMIB uses a conditional multimodal information bottleneck to create reusable agent skills that separate verbalizable text content from predictive perceptual residuals, improving execution stability.
InvEvolve evolves white-box inventory policies from LLMs with statistical safety guarantees and outperforms classical and deep learning methods on synthetic and real retail data.
o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
ASB is a new benchmark that tests 10 prompt injection attacks, memory poisoning, a novel Plan-of-Thought backdoor attack, and 11 defenses on LLM agents across 13 models, finding attack success rates up to 84.3% and limited defense effectiveness.
Activation Addition steers language models by adding contrastive activation vectors from prompt pairs to control high-level properties like sentiment and toxicity at inference time without training.
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
BoolXLLM augments an existing Boolean rule learner with LLMs for feature selection, discretization thresholds, and natural-language rule translation to improve interpretability while preserving accuracy.
Interpretability research should be judged by actionability—the degree to which its insights support concrete decisions and interventions—rather than explanatory power alone.
Agreement-based clustering of annotators improves performance on subjective NLP tasks by capturing diverse perspectives better than majority voting or per-annotator modeling.
HAGE proposes a trainable weighted graph memory framework with LLM intent classification, dynamic edge modulation, and RL optimization that improves long-horizon reasoning accuracy in agentic LLMs over static baselines.
PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.
APCD reduces LLM hallucinations by expanding decoding paths adaptively when entropy signals uncertainty and by contrasting divergent paths to control their interaction.
RKU is a curvature-aware structural pruning framework that improves LLM reasoning accuracy at 40% sparsity, reaching 13.34% on GSM8K while outperforming baselines and better preserving out-of-distribution representations.
Expanded recall in LLM agents erodes cooperative intent in multi-agent social dilemmas, observed in 18 of 28 model-game settings.
PaT defers planning until after failed trials in LLM code generation, enabling heterogeneous cheap-plus-powerful model setups that match large-model performance at roughly 69% lower cost.
BoundaryRouter routes queries to LLM or agent using early experience memory from a seed set, cutting inference time 60.6% versus always using agents and raising performance 28.6% versus always using direct LLM inference.
APPS approximates power targets p(x)^alpha via parallel particle propagation with proposal-corrected reweighting and future-value-guided selection at block boundaries, improving accuracy-runtime trade-offs in training-free decoding.
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
Muon optimizer with weight decay and update scaling achieves ~2x efficiency over AdamW for large LLMs, shown via the Moonlight 3B/16B MoE model trained on 5.7T tokens.
HSUGA improves LLM-enhanced sequential recommendation via staged hierarchical semantic understanding for better preference extraction and group-aware alignment that varies intensity by user activity level.
RADAR is a redundancy-aware, query-adaptive framework that uses conditional discrete graph diffusion to generate efficient communication topologies for multi-agent LLM systems, outperforming baselines on six benchmarks with higher accuracy and lower token use.
citing papers explorer
-
Why Users Go There: World Knowledge-Augmented Generative Next POI Recommendation
AWARE augments generative next-POI recommendation with LLM agents that produce user-anchored narratives capturing events, culture, and trends, delivering up to 12.4% relative gains on three real datasets.
-
OLIVIA: Online Learning via Inference-time Action Adaptation for Decision Making in LLM ReAct Agents
OLIVIA treats LLM agent action selection as a contextual linear bandit over frozen hidden states and applies UCB exploration to adapt online, yielding consistent gains over static ReAct and prompt-based baselines on four benchmarks.
-
Fin-Bias: Comprehensive Evaluation for LLM Decision-Making under human bias in Finance Domain
LLMs copy biased analyst ratings in investment decisions but a new detection method encourages independent reasoning and can improve stock return predictions beyond human levels.
-
Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck
CMIB uses a conditional multimodal information bottleneck to create reusable agent skills that separate verbalizable text content from predictive perceptual residuals, improving execution stability.
-
InvEvolve: Evolving White-Box Inventory Policies via Large Language Models with Performance Guarantees
InvEvolve evolves white-box inventory policies from LLMs with statistical safety guarantees and outperforms classical and deep learning methods on synthetic and real retail data.
-
Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
o1-like models overthink easy tasks; self-training reduces compute use without accuracy loss on GSM8K, MATH500, GPQA, and AIME.
-
Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents
ASB is a new benchmark that tests 10 prompt injection attacks, memory poisoning, a novel Plan-of-Thought backdoor attack, and 11 defenses on LLM agents across 13 models, finding attack success rates up to 84.3% and limited defense effectiveness.
-
Steering Language Models With Activation Engineering
Activation Addition steers language models by adding contrastive activation vectors from prompt pairs to control high-level properties like sentiment and toxicity at inference time without training.
-
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
-
BoolXLLM: LLM-Assisted Explainability for Boolean Models
BoolXLLM augments an existing Boolean rule learner with LLMs for feature selection, discretization thresholds, and natural-language rule translation to improve interpretability while preserving accuracy.
-
Interpretability Can Be Actionable
Interpretability research should be judged by actionability—the degree to which its insights support concrete decisions and interventions—rather than explanatory power alone.
-
Beyond Majority Voting: Agreement-Based Clustering to Model Annotator Perspectives in Subjective NLP Tasks
Agreement-based clustering of annotators improves performance on subjective NLP tasks by capturing diverse perspectives better than majority voting or per-annotator modeling.
-
HAGE: Harnessing Agentic Memory via RL-Driven Weighted Graph Evolution
HAGE proposes a trainable weighted graph memory framework with LLM intent classification, dynamic edge modulation, and RL optimization that improves long-horizon reasoning accuracy in agentic LLMs over static baselines.
-
PruneTIR: Inference-Time Tool Call Pruning for Effective yet Efficient Tool-Integrated Reasoning
PruneTIR prunes erroneous tool-call trajectories during LLM inference via three trigger-based components to raise Pass@1 accuracy and efficiency while shortening context.
-
APCD: Adaptive Path-Contrastive Decoding for Reliable Large Language Model Generation
APCD reduces LLM hallucinations by expanding decoding paths adaptively when entropy signals uncertainty and by contrasting divergent paths to control their interaction.
-
Relative Kinetic Utility for Reasoning-Aware Structural Pruning in Large Language Models
RKU is a curvature-aware structural pruning framework that improves LLM reasoning accuracy at 40% sparsity, reaching 13.34% on GSM8K while outperforming baselines and better preserving out-of-distribution representations.
-
The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agents
Expanded recall in LLM agents erodes cooperative intent in multi-agent social dilemmas, observed in 18 of 28 model-game settings.
-
PaT: Planning-after-Trial for Efficient Test-Time Code Generation
PaT defers planning until after failed trials in LLM code generation, enabling heterogeneous cheap-plus-powerful model setups that match large-model performance at roughly 69% lower cost.
-
Learning Agent Routing From Early Experience
BoundaryRouter routes queries to LLM or agent using early experience memory from a seed set, cutting inference time 60.6% versus always using agents and raising performance 28.6% versus always using direct LLM inference.
-
The Model Knows, the Decoder Finds: Future Value Guided Particle Power Sampling
APPS approximates power targets p(x)^alpha via parallel particle propagation with proposal-corrected reweighting and future-value-guided selection at block boundaries, improving accuracy-runtime trade-offs in training-free decoding.
-
Towards an AI co-scientist
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
-
Muon is Scalable for LLM Training
Muon optimizer with weight decay and update scaling achieves ~2x efficiency over AdamW for large LLMs, shown via the Moonlight 3B/16B MoE model trained on 5.7T tokens.
-
HSUGA: LLM-Enhanced Recommendation with Hierarchical Semantic Understanding and Group-Aware Alignment
HSUGA improves LLM-enhanced sequential recommendation via staged hierarchical semantic understanding for better preference extraction and group-aware alignment that varies intensity by user activity level.
-
RADAR: Redundancy-Aware Diffusion for Multi-Agent Communication Structure Generation
RADAR is a redundancy-aware, query-adaptive framework that uses conditional discrete graph diffusion to generate efficient communication topologies for multi-agent LLM systems, outperforming baselines on six benchmarks with higher accuracy and lower token use.
-
Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning
Skill1 trains a single RL policy to co-evolve skill selection, utilization, and distillation in language model agents from one task-outcome reward, using low-frequency trends to credit selection and high-frequency variation to credit distillation, outperforming baselines on ALFWorld and WebShop.
-
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Multi-agent debate with tit-for-tat arguments and a judge LLM improves reasoning by preventing LLMs from locking into incorrect initial solutions.
-
VulTriage: Triple-Path Context Augmentation for LLM-Based Vulnerability Detection
VulTriage combines control dependency extraction, CWE knowledge retrieval, and semantic summarization to improve LLM accuracy on vulnerability detection, reaching SOTA on PrimeVul and generalizing to Kotlin.