OSA improves LLM-based recommenders by anchoring ordinal preference levels as numeric tokens in the model's latent space to retain fine-grained strength information when fusing collaborative signals.
Mixed citations
Title resolution pending
Mixed citation behavior. Most common role is method (60%).
37
Pith papers citing it
Method
60% of classified citations
citation-role summary
method 3
background 2
citation-polarity summary
representative citing papers
RLFSeg repurposes pretrained generative models via Rectified Flow for direct latent-space image-to-mask mapping in text-based segmentation, outperforming diffusion-based methods especially in zero-shot cases.
A product-key parametric memory head with selective sparse updates mitigates catastrophic forgetting in generative retrieval models during sequential addition of new documents.
SMTPO uses multi-task SFT to improve simulator feedback quality and RL with fine-grained rewards to optimize multi-turn preference reasoning in LLM-based conversational recommendation.
HeiSD delivers up to 2.45x faster inference for embodied VLA models by hybridizing speculative decoding with kinematic boundary detection and error-mitigation tricks while preserving task success rates.
EvoESAP uses evolutionary search guided by a speculative-decoding-inspired ESAP metric to discover non-uniform layer-wise sparsity allocations for MoE expert pruning, improving generation accuracy up to 19.6% at 50% sparsity.
ProAgent uses on-demand tiered perception and context-aware LLM reasoning to deliver proactive assistance on AR glasses, achieving up to 27.7% higher prediction accuracy and 20.5% lower false detections than baselines.
Optimal preference elicitation in conversational recommenders is stage-dependent (attributes early, items later), and a MoE model trained on a new annotated dataset improves offline recommendation and response quality.
EPnG reallocates LoRA capacity in MoE models by pruning experts with low router gate probabilities and expanding high-importance ones via rank growth, outperforming standard LoRA and nearing full fine-tuning performance with 0.55-0.72% parameters updated.
CoPersona introduces a multiplex persona graph for facet-level peer alignment and a dual-branch retrieval-plus-reasoning architecture to improve LLM personalization under sparse and biased user interaction data.
REBench is a new benchmark that consolidates existing datasets into a large collection of binaries with knowledge-base-driven ground truth to enable fair LLM evaluation on stripped-binary type and name recovery.
A scalable training-free pipeline using video segmentation, filtering, and off-the-shelf multimodal models creates DenseStep2M, a dataset of 100K videos and 2M detailed instructional steps that improves dense captioning, step grounding, and cross-modal retrieval.
DC4SR improves sequential recommendation denoising by iteratively calibrating LLM semantic priors and model learning posteriors using their disagreement as a signal for better alignment with true user interests.
ScanVLA uses a vision-language model with a history-enhanced decoder and frozen segmentation LoRA to outperform prior methods on object-referring scanpath prediction.
LWGR applies personalized soft instructions for LLM knowledge extraction and Lagrangian primal-dual optimization to selectively fuse beneficial world knowledge into generative recommendation while bounding degradation.
SpotSound adds a hallucination-suppressing objective and a needle-in-haystack benchmark to audio-language models, reaching state-of-the-art temporal grounding while keeping general task performance.
UniDetect is an LLM-based system that generates universal transaction summary texts and uses two-stage multimodal training on text plus graphs to detect fraudulent accounts across heterogeneous blockchains, outperforming baselines by 5.57-7.58% KS and achieving over 94.58% zero-shot cross-chain and
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
KnowSA_CKP uses comparative knowledge probing to selectively augment LLM prompts for items with knowledge gaps, improving recommendation accuracy and context efficiency.
UATTA adapts pre-trained text-image models at test time without labels by using disagreement in bidirectional retrieval rankings to estimate and mitigate uncertainty for improved person search.
CRAB mitigates popularity bias in generative recommenders by rebalancing the semantic token codebook through splitting popular tokens and applying a tree-structured regularizer to boost representations for unpopular items.
OpsLLM's pipeline (HITL data curation, SFT, GRPO RL with a domain process reward model) improves LLM accuracy on software-operations QA and RCA, especially on in-distribution root-cause-analysis tasks.
CKG-LLM uses LLMs to generate executable queries over contract knowledge graphs for detecting access control vulnerabilities and reports superior performance versus existing tools.
MAP4TS combines global, local, statistical, and temporal prompts derived from classical time-series analysis with raw embeddings via cross-modality alignment to improve LLM forecasting performance across eight datasets.
citing papers explorer
-
Every Preference Has Its Strength: Injecting Ordinal Semantics into LLM-Based Recommenders
OSA improves LLM-based recommenders by anchoring ordinal preference levels as numeric tokens in the model's latent space to retain fine-grained strength information when fusing collaborative signals.
-
From Diffusion to Rectified Flow: Rethinking Text-Based Segmentation
RLFSeg repurposes pretrained generative models via Rectified Flow for direct latent-space image-to-mask mapping in text-based segmentation, outperforming diffusion-based methods especially in zero-shot cases.
-
A Parametric Memory Head for Continual Generative Retrieval
A product-key parametric memory head with selective sparse updates mitigates catastrophic forgetting in generative retrieval models during sequential addition of new documents.
-
User Simulator-Guided Multi-Turn Preference Optimization for Reasoning LLM-based Conversational Recommendation
SMTPO uses multi-task SFT to improve simulator feedback quality and RL with fine-grained rewards to optimize multi-turn preference reasoning in LLM-based conversational recommendation.
-
HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness
HeiSD delivers up to 2.45x faster inference for embodied VLA models by hybridizing speculative decoding with kinematic boundary detection and error-mitigation tricks while preserving task success rates.
-
EvoESAP: Non-Uniform Expert Pruning for Sparse MoE
EvoESAP uses evolutionary search guided by a speculative-decoding-inspired ESAP metric to discover non-uniform layer-wise sparsity allocations for MoE expert pruning, improving generation accuracy up to 19.6% at 50% sparsity.
-
ProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems in the Wild
ProAgent uses on-demand tiered perception and context-aware LLM reasoning to deliver proactive assistance on AR glasses, achieving up to 27.7% higher prediction accuracy and 20.5% lower false detections than baselines.
-
When and How to Ask: Dynamic Preference Elicitation Strategies for Conversational Recommendation
Optimal preference elicitation in conversational recommenders is stage-dependent (attributes early, items later), and a MoE model trained on a new annotated dataset improves offline recommendation and response quality.
-
EPnG: Adaptive Expert Prune-and-Grow for Parameter-Efficient MoE Fine-tuning
EPnG reallocates LoRA capacity in MoE models by pruning experts with low router gate probabilities and expanding high-importance ones via rank growth, outperforming standard LoRA and nearing full fine-tuning performance with 0.55-0.72% parameters updated.
-
CoPersona: Collaborative Persona Graphs for Robust LLM Personalization
CoPersona introduces a multiplex persona graph for facet-level peer alignment and a dual-branch retrieval-plus-reasoning architecture to improve LLM personalization under sparse and biased user interaction data.
-
REBENCH: A Procedural, Fair-by-Construction Benchmark for LLMs on Stripped-Binary Types and Names (Extended Version)
REBench is a new benchmark that consolidates existing datasets into a large collection of binaries with knowledge-base-driven ground truth to enable fair LLM evaluation on stripped-binary type and name recovery.
-
DenseStep2M: A Scalable, Training-Free Pipeline for Dense Instructional Video Annotation
A scalable training-free pipeline using video segmentation, filtering, and off-the-shelf multimodal models creates DenseStep2M, a dataset of 100K videos and 2M detailed instructional steps that improves dense captioning, step grounding, and cross-modal retrieval.
-
Disagreement as Signals: Dual-view Calibration for Sequential Recommendation Denoising
DC4SR improves sequential recommendation denoising by iteratively calibrating LLM semantic priors and model learning posteriors using their disagreement as a signal for better alignment with true user interests.
-
Object Referring-Guided Scanpath Prediction with Perception-Enhanced Vision-Language Models
ScanVLA uses a vision-language model with a history-enhanced decoder and frozen segmentation LoRA to outperform prior methods on object-referring scanpath prediction.
-
LWGR: Lagrangian-Constrained Personalized World Knowledge for Generative Recommendation
LWGR applies personalized soft instructions for LLM knowledge extraction and Lagrangian primal-dual optimization to selectively fuse beneficial world knowledge into generative recommendation while bounding degradation.
-
SpotSound: Enhancing Large Audio-Language Models with Fine-Grained Temporal Grounding
SpotSound adds a hallucination-suppressing objective and a needle-in-haystack benchmark to audio-language models, reaching state-of-the-art temporal grounding while keeping general task performance.
-
UniDetect: LLM-Driven Universal Fraud Detection across Heterogeneous Blockchains
UniDetect is an LLM-based system that generates universal transaction summary texts and uses two-stage multimodal training on text plus graphs to detect fraudulent accounts across heterogeneous blockchains, outperforming baselines by 5.57-7.58% KS and achieving over 94.58% zero-shot cross-chain and
-
ARHN: Answer-Centric Relabeling of Hard Negatives with Open-Source LLMs for Dense Retrieval
ARHN refines hard-negative training data for dense retrieval by using LLMs to convert answer-containing passages into additional positives and exclude answer-containing passages from the negative set.
-
Filling the Gaps: Selective Knowledge Augmentation for LLM Recommenders
KnowSA_CKP uses comparative knowledge probing to selectively augment LLM prompts for items with knowledge gaps, improving recommendation accuracy and context efficiency.
-
Pretrain-then-Adapt: Uncertainty-Aware Test-Time Adaptation for Text-based Person Search
UATTA adapts pre-trained text-image models at test time without labels by using disagreement in bidirectional retrieval rankings to estimate and mitigate uncertainty for improved person search.
-
CRAB: Codebook Rebalancing for Bias Mitigation in Generative Recommendation
CRAB mitigates popularity bias in generative recommenders by rebalancing the semantic token codebook through splitting popular tokens and applying a tree-structured regularizer to boost representations for unpopular items.
-
OpsLLM: Construction of Large Language Model for Software Operations with Multi-stage Learning
OpsLLM's pipeline (HITL data curation, SFT, GRPO RL with a domain process reward model) improves LLM accuracy on software-operations QA and RCA, especially on in-distribution root-cause-analysis tasks.
-
CKG-LLM: LLM-Assisted Detection of Smart Contract Access Control Vulnerabilities Based on Knowledge Graphs
CKG-LLM uses LLMs to generate executable queries over contract knowledge graphs for detecting access control vulnerabilities and reports superior performance versus existing tools.
-
MAP4TS: A Multi-Aspect Prompting Framework for Time-Series Forecasting with Large Language Models
MAP4TS combines global, local, statistical, and temporal prompts derived from classical time-series analysis with raw embeddings via cross-modality alignment to improve LLM forecasting performance across eight datasets.
-
MindAdapter: Few-Shot Parameter-Efficient Residual Calibration of Cross-Subject Brain-to-Visual Decoding Models
MindAdapter introduces a decoupled linear-residual adapter with topology-anchored dual-stream manifold constraints for few-shot cross-subject calibration of brain-to-visual models, improving reconstruction and retrieval on NSD.
-
VaaWIT: Visual-Aware Adaptation of Large Language Models for Multilingual Web Image Translation
VaaWIT proposes DSAM and VAA modules to adapt LLMs for multilingual web image translation, claiming outperformance over open-source baselines on benchmarks.
-
Boosting Automatic Java-to-Cangjie Translation with Multi-Stage LLM Training and Error Repair
Multi-stage LLM training plus compiler-guided error repair boosts functional equivalence in Java-to-Cangjie translation by 6.06% over prior methods despite scarce parallel data.
-
HiMix: Hierarchical Artifact-aware Mixup for Generalized Synthetic Image Detection
HiMix combines mixup augmentation to create transitional real-fake samples with hierarchical global-local artifact feature fusion to achieve better generalization in detecting AI-generated images from unseen generators.
-
Frequency-Aware Semantic Fusion with Gated Injection for AI-generated Image Detection
FGINet uses a band-masked frequency encoder and layer-wise gated injection to fuse frequency artifacts with vision foundation model semantics, plus hyperspherical compactness learning, to achieve better generalization in AI-generated image detection.
-
SAGE: A Strategy-Aware Graph-Enhanced Generation Framework For Online Counseling
SAGE uses a Next Strategy Classifier and Graph-Aware Attention on a psychologically grounded graph to improve LLM strategy prediction and response quality in online counseling.
-
Discrete Preference Learning for Personalized Multimodal Generation
DPPMG learns discrete modal-specific preferences via a dedicated GNN from multimodal user data, quantizes them into tokens, and feeds them into generators with a consistency reward to produce personalized text and images.
-
A Utility-preserving De-identification Pipeline for Cross-hospital Radiology Data Sharing
The UPDP pipeline filters privacy terms and generates de-identified radiology images that preserve diagnostic pathology information, enabling models with competitive disease detection accuracy but reduced identity leakage and improved cross-hospital performance.
-
WRF4CIR: Weight-Regularized Fine-Tuning Network for Composed Image Retrieval
WRF4CIR uses weight-regularized fine-tuning with adversarial perturbations to mitigate overfitting in composed image retrieval and narrows the generalization gap on benchmarks.
-
Structural and Disentangled Adaptation of Large Vision Language Models for Multimodal Recommendation
SDA uses structural alignment as a soft teacher and gated low-rank expert paths to adapt LVLMs for multimodal recommendation, reporting 6.15% Hit@10 and 8.64% NDCG@10 average gains plus larger long-tail improvements on Amazon datasets.
-
SHE: Stepwise Hybrid Examination Reinforcement Learning Framework for E-commerce Search Relevance
SHE is a new RL framework using stepwise hybrid examination rewards to improve reasoning quality and accuracy in large-scale e-commerce query-product relevance prediction.
-
CurEvo: Curriculum-Guided Self-Evolution for Video Understanding
CurEvo integrates curriculum guidance into self-evolution to structure autonomous improvement of video understanding models, yielding gains on VideoQA benchmarks.
-
Data-Driven Function Calling Improvements in Large Language Model for Online Financial QA
A pipeline of dataset construction from prior work, AugFC parameter augmentation, and two-step LLM training improves function calling for financial APIs and is running in production.