Prompt injection attacks can self-replicate across LLM agents in multi-agent systems, enabling data theft, misinformation, and system disruption while propagating silently.
hub Mixed citations
Nature commu- nications, 13(1):862
Mixed citation behavior. Most common role is background (57%).
31
Pith papers citing it
98
external citations · external index
Background
57% of classified citations
hub tools
citation-role summary
background 5
method 1
other 1
citation-polarity summary
representative citing papers
Doc-to-Atom decomposes documents into composable micro-LoRA adapters selected by a query router for efficient long-context QA.
A cycle-consistent MT pipeline generates and similarity-weights training data for coreference resolution, producing gains on four low-resource languages and enabling the task where no corpora existed.
StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
SAri-RFT applies GRPO-based reinforcement fine-tuning to LVLMs on novel two-term and three-term visual semantic arithmetic tasks, reaching SOTA on the new IRPD dataset and Visual7W-Telling.
Users treat human delegation for long tasks as a flexible compass but AI delegation as rigid railway tracks due to perceived AI limitations in inference and judgment.
Low-rank LLM adaptation during vision-language alignment outperforms full fine-tuning by preserving per-token visual structure and favoring flat, noise-robust subspaces.
PRIME is a new evaluation framework that creates calibrated conflicts in LLM prompts and finds conflict type affects model behavior more than scale.
SafeSteer restricts reverse KL penalty to safety tokens selected via activation steering, achieving strong safety on seven benchmarks with minimal degradation on five capability benchmarks using only 100 harmful samples and no general data.
Introduces Layout-as-Policy (LaP) to turn 3D layout estimation into an iterative policy-learning refinement process for better physical coherence.
A dataset-agnostic framework converts text tool-calling benchmarks to paired audio evaluations via TTS, speaker variation and noise, then evaluates seven omni-modal models showing model- and task-dependent performance with small text-to-voice gaps.
DKPS-based methods predict new model benchmark scores using cached responses, matching baseline mean absolute error with substantially fewer queries and an offline query selection approach.
NeWTral is a non-linear weight translation framework using MoE routing that reduces average attack success rate from 70% to 13% on unsafe domain adapters across Llama, Mistral, Qwen, and Gemma models up to 72B while retaining 90% knowledge fidelity.
CURE-MED pairs a new 13-language medical reasoning benchmark with curriculum RL to raise logical correctness to 70% and language consistency to 95% at 32B scale while outperforming baselines.
MoE-LLaVA applies mixture-of-experts sparsity to LVLMs via MoE-Tuning, delivering LLaVA-1.5-7B level visual understanding and better hallucination resistance with only ~3B active parameters.
Curating 25K natural-instruction examples for RLVR yields a 64.4pp relative gain on BBEH for Qwen3-0.6B and generalizes across scales and model families.
Survey mapping RL techniques onto LLM training and highlighting gaps in value-based, off-policy, and bootstrapping methods.
SEEK uses adaptive semantic chunking to create complete evidence units and fine-tunes multilingual LLMs with LoRA, achieving up to 20% better macro-F1 on fact-checking datasets compared to baselines.
A sparse-autoencoder-based, label-free steering update to CLIP image embeddings improves zero-shot classification on nine datasets and comes with an FVU-based reliability gate.
The survey organizes causes of hallucinations in MLLMs, reviews evaluation benchmarks and metrics, and outlines mitigation approaches plus open questions.
Distilling CoT from DeepSeek-R1 to Qwen2.5-7B on competition problems yields 4.76 pp accuracy gain to 69.43% and 73.1% on MATH-500, with accuracy falling as response length decreases.
VisShield with OPTIC dataset enables VLMs to localize and mask private text in vision data via instruction tuning for privacy preservation.
A CatBoost regressor predicts response length before D-LLM generation, cutting padded-token compute while claiming preserved output quality.
Decomposing automotive query understanding into a lightweight classification stage followed by specialized entity extraction yields better accuracy and lower latency than joint single-step processing.
citing papers explorer
-
Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems
Prompt injection attacks can self-replicate across LLM agents in multi-agent systems, enabling data theft, misinformation, and system disruption while propagating silently.
-
Doc-to-Atom: Learning to Compile and Compose Memory Atoms
Doc-to-Atom decomposes documents into composable micro-LoRA adapters selected by a query router for efficient long-context QA.
-
Multilingual Coreference Resolution via Cycle-Consistent Machine Translation
A cycle-consistent MT pipeline generates and similarity-weights training data for coreference resolution, producing gains on four low-resource languages and enabling the task where no corpora existed.
-
StepCodeReasoner: Aligning Code Reasoning with Stepwise Execution Traces via Reinforcement Learning
StepCodeReasoner aligns code reasoning with verifiable stepwise execution traces via print anchors and bi-level GRPO reinforcement learning, reaching SOTA results on CRUXEval (91.1%) and LiveCodeBench (86.5%) for a 7B model.
-
Multi-modal Reasoning with LLMs for Visual Semantic Arithmetic
SAri-RFT applies GRPO-based reinforcement fine-tuning to LVLMs on novel two-term and three-term visual semantic arithmetic tasks, reaching SOTA on the new IRPD dataset and Visual7W-Telling.
-
Compass vs Railway Tracks: Unpacking User Mental Models for Communicating Long-Horizon Work to Humans vs. AI
Users treat human delegation for long tasks as a flexible compass but AI delegation as rigid railway tracks due to perceived AI limitations in inference and judgment.
-
Dive Into the Implicit Biases of Low-rank Vision-language Alignment
Low-rank LLM adaptation during vision-language alignment outperforms full fine-tuning by preserving per-token visual structure and favoring flat, noise-robust subspaces.
-
PRIME: Evaluating Prompt Resolution Under Incompatible Instructions in LLMs
PRIME is a new evaluation framework that creates calibrated conflicts in LLM prompts and finds conflict type affects model behavior more than scale.
-
SafeSteer: Localized On-Policy Distillation for Efficient Safety Alignment
SafeSteer restricts reverse KL penalty to safety tokens selected via activation steering, achieving strong safety on seven benchmarks with minimal degradation on five capability benchmarks using only 100 harmful samples and no general data.
-
Perceive-then-Plan: Layout-as-Policy for Monocular 3D Scene Layout Estimation
Introduces Layout-as-Policy (LaP) to turn 3D layout estimation into an iterative policy-learning refinement process for better physical coherence.
-
From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents
A dataset-agnostic framework converts text tool-calling benchmarks to paired audio evaluations via TTS, speaker variation and noise, then evaluates seven omni-modal models showing model- and task-dependent performance with small text-to-voice gaps.
-
Query-efficient model evaluation using cached responses
DKPS-based methods predict new model benchmark scores using cached responses, matching baseline mean absolute error with substantially fewer queries and an offline query selection approach.
-
You Snooze, You Lose: Automatic Safety Alignment Restoration through Neural Weight Translation
NeWTral is a non-linear weight translation framework using MoE routing that reduces average attack success rate from 70% to 13% on unsafe domain adapters across Llama, Mistral, Qwen, and Gemma models up to 72B while retaining 90% knowledge fidelity.
-
CURE-Med: Curriculum-Informed Reinforcement Learning for Multilingual Medical Reasoning
CURE-MED pairs a new 13-language medical reasoning benchmark with curriculum RL to raise logical correctness to 70% and language consistency to 95% at 32B scale while outperforming baselines.
-
MoE-LLaVA: Mixture of Experts for Large Vision-Language Models
MoE-LLaVA applies mixture-of-experts sparsity to LVLMs via MoE-Tuning, delivering LLaVA-1.5-7B level visual understanding and better hallucination resistance with only ~3B active parameters.
-
SUPERNOVA: Eliciting General Reasoning in LLMs with Reinforcement Learning on Natural Instructions
Curating 25K natural-instruction examples for RLVR yields a 64.4pp relative gain on BBEH for Qwen3-0.6B and generalizes across scales and model families.
-
Modularized Reinforcement Learning on LLMs: From MDP Creation to Exploration and Learning
Survey mapping RL techniques onto LLM training and highlighting gaps in value-based, off-policy, and bootstrapping methods.
-
SEEK: Semantic Evidence Extraction via Adaptive ChunKing for Multilingual Fact-Checking
SEEK uses adaptive semantic chunking to create complete evidence units and fine-tunes multilingual LLMs with LoRA, achieving up to 20% better macro-F1 on fact-checking datasets compared to baselines.
-
Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering
A sparse-autoencoder-based, label-free steering update to CLIP image embeddings improves zero-shot classification on nine datasets and comes with an FVU-based reliability gate.
-
Hallucination of Multimodal Large Language Models: A Survey
The survey organizes causes of hallucinations in MLLMs, reviews evaluation benchmarks and metrics, and outlines mitigation approaches plus open questions.
-
Knowledge Distillation from Large Reasoning Models to Compact Student Models: A Case Study on the John O Bryan Mathematics Competition
Distilling CoT from DeepSeek-R1 to Qwen2.5-7B on competition problems yields 4.76 pp accuracy gain to 69.43% and 73.1% on MATH-500, with accuracy falling as response length decreases.
-
Vision Language Model Helps Private Information De-Identification in Vision Data
VisShield with OPTIC dataset enables VLMs to localize and mask private text in vision data via instruction tuning for privacy preservation.
-
Predict-then-Diffuse: Adaptive Response Length for Compute-Budgeted Inference in Diffusion LLMs
A CatBoost regressor predicts response length before D-LLM generation, cutting padded-token compute while claiming preserved output quality.
-
Domain-Specific Query Understanding for Automotive Applications: A Modular and Scalable Approach
Decomposing automotive query understanding into a lightweight classification stage followed by specialized entity extraction yields better accuracy and lower latency than joint single-step processing.
-
Less LLM, More Documents: Searching for Improved RAG
Corpus scaling in RAG frequently matches the accuracy gains from larger LLMs on open-domain QA tasks, with mid-sized models benefiting most due to better passage coverage.
-
Large Language Models for Multi-Robot Systems: A Survey
A survey that categorizes LLM uses in multi-robot systems across task allocation, motion planning, action generation, and human interaction, while noting challenges and future research opportunities.
-
Large Language Model-Brained GUI Agents: A Survey
A survey consolidating frameworks, data practices, large action models, benchmarks, applications, and research gaps in LLM-brained GUI agents.
-
Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
The paper surveys reinforced reasoning techniques for LLMs, covering automated data construction, learning-to-reason methods, and test-time scaling as steps toward Large Reasoning Models.
-
A Survey on the Memory Mechanism of Large Language Model based Agents
A systematic review of memory designs, evaluation methods, applications, limitations, and future directions for LLM-based agents.
- Investigating Adversarial Robustness of Multi-modal Large Language Models
- Property Enhanced Instruction Tuning for Multi-task Molecule Generation with Large Language Models