SeedHijack is a blind, integrity-preserving PRNG hijacking attack that amplifies LLM watermark z-scores up to 2.42x while evading all tested content-side statistical detectors across three schemes and models.
super hub Mixed citations
Qwen Technical Report
Mixed citation behavior. Most common role is background (67%).
436
Pith papers citing it
Background
67% of classified citations
abstract
Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural language processing tasks that were previously thought to be exclusive to humans. In this work, we introduce Qwen, the first installment of our large language model series. Qwen is a comprehensive language model series that encompasses distinct models with varying parameter counts. It includes Qwen, the base pretrained language models, and Qwen-Chat, the chat models finetuned with human alignment techniques. The base language models consistently demonstrate superior performance across a multitude of downstream tasks, and the chat models, particularly those trained using Reinforcement Learning from Human Feedback (RLHF), are highly competitive. The chat models possess advanced tool-use and planning capabilities for creating agent applications, showcasing impressive performance even when compared to bigger models on complex tasks like utilizing a code interpreter. Furthermore, we have developed coding-specialized models, Code-Qwen and Code-Qwen-Chat, as well as mathematics-focused models, Math-Qwen-Chat, which are built upon base language models. These models demonstrate significantly improved performance in comparison with open-source models, and slightly fall behind the proprietary models.
hub tools
citation-role summary
background 82
baseline 16
method 16
dataset 1
extension 1
other 1
citation-polarity summary
claims ledger
- abstract Large language models (LLMs) have revolutionized the field of artificial intelligence, enabling natural language processing tasks that were previously thought to be exclusive to humans. In this work, we introduce Qwen, the first installment of our large language model series. Qwen is a comprehensive language model series that encompasses distinct models with varying parameter counts. It includes Qwen, the base pretrained language models, and Qwen-Chat, the chat models finetuned with human alignment techniques. The base language models consistently demonstrate superior performance across a mult
authors
co-cited works
representative citing papers
LongAct benchmark evaluates long-horizon household task execution from free-form instructions; HoloMind agent raises performance but top VLMs still reach only 59% goal completion and 16% full-task success.
The upper-tail accumulation scale derived from the gap-counting function N_n sets the critical inverse temperature for softmax attention concentration, unifying prior conflicting laws as special cases of different N_n.
IRIS-14B is the first LLM trained explicitly for GIMPLE-to-LLVM IR translation and outperforms much larger models by up to 44 percentage points on real-world C code.
A classifier trained only on transformer fine-tuning data detects an invariant memorization signature that transfers to Mamba, RWKV-4, and RecurrentGemma with AUCs of 0.963, 0.972, and 0.936.
LiveBench is a contamination-limited LLM benchmark with auto-scored challenging tasks from recent sources across math, coding, reasoning and more, where top models score below 70%.
OSWorld provides the first unified real-computer benchmark for open-ended multimodal agent tasks, exposing large performance gaps between humans and state-of-the-art LLM/VLM agents.
A new sensitivity-labeled test collection is released from Enron emails with crowdsourced queries, relevance judgments, and LLM extensions for evaluating sensitivity-aware search.
Moral Trolley Arena shows frontier LLMs produce composite moral preferences that are compressed rather than additive functions of calibrated component act strengths across Moral Foundations Theory.
Introduces ChiSafe-PAS, a 1,897-prompt human-annotated Chinese adversarial benchmark for LLM safety with 3-class labels, 9-category obfuscation taxonomy, and domain coverage in self-harm, drugs, fraud, and satire.
EvoGM uses a dual-generator architecture with cycle-consistent learning on winner-loser pairs from search history to optimize LLM merging coefficients inside a multi-round evolutionary pipeline and reports outperformance over baselines on seen and unseen tasks.
VisAnalog is a new controlled benchmark showing VLMs substantially underperform humans on visual concept transfer under one- to four-step deterministic transformations, with relation inference as the main failure mode.
Proposes an equation-anchored tool-use method for MLLMs that writes the pinhole back-projection equation in Chain-of-Thought and substitutes retrieved camera intrinsics and depths to achieve robustness in 3D object detection and visual grounding under rescaled intrinsics.
Incantation is the first video world model to use per-frame natural language conditioning for simultaneous multi-entity control and concept-level cross-entity transfer in interactive video generation.
TIDAL recovers temporal phase signals from LLM-derived semantics of provisioning metadata to enable complementary CVD placement, reducing overload frequency by 79.1% on production traces.
RISE is an inference-time semantic reranking framework that refines low-confidence predictions in rhetorical role labeling using contrastively learned label representations, delivering an average +9.15 macro-F1 gain on hard examples across eight datasets and seven models.
DCDM replaces positional blocks with learnable semantic chunks via differentiable Chunking Attention, yielding consistent gains over block and unstructured diffusion baselines up to 1.5B parameters.
ClawForge is a generator framework that creates reproducible executable benchmarks for command-line agents under state conflict, with ClawForge-Bench showing frontier models reach at most 45.3% strict accuracy and that state inspection drives most performance gaps.
BlockVLA accelerates autoregressive VLA models by 3.3x using block diffusion finetuning, with faster training convergence and better early performance on long-horizon robotic tasks.
Pretrained LLMs adapted via convolutional projections and LoRA act as efficient frozen backbones for sensor-based human activity recognition, delivering strong data efficiency and cross-dataset transfer.
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
V-ABS is an action-observer beam search method with entropy-based adaptive weighting and an 80k-sample SFT dataset that delivers 19.7% average gains on visual reasoning tasks for MLLMs.
LLM agents reach only 50.6% accuracy on chemical cost estimation within 25% error even with tools, dropping with noise due to parsing, pack selection, and tool-use failures.
DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
citing papers explorer
-
Blind PRNG Hijacking: An Undetectable Integrity-Preserving Attack Against LLM Watermarking
SeedHijack is a blind, integrity-preserving PRNG hijacking attack that amplifies LLM watermark z-scores up to 2.42x while evading all tested content-side statistical detectors across three schemes and models.
-
When Robots Do the Chores: A Benchmark and Agent for Long-Horizon Household Task Execution
LongAct benchmark evaluates long-horizon household task execution from free-form instructions; HoloMind agent raises performance but top VLMs still reach only 59% goal completion and 16% full-task success.
-
A Unified Framework for Critical Scaling of Inverse Temperature in Self-Attention
The upper-tail accumulation scale derived from the gap-counting function N_n sets the critical inverse temperature for softmax attention concentration, unifying prior conflicting laws as special cases of different N_n.
-
LLM Translation of Compiler Intermediate Representation
IRIS-14B is the first LLM trained explicitly for GIMPLE-to-LLVM IR translation and outperforms much larger models by up to 44 percentage points on real-world C code.
-
Learning the Signature of Memorization in Autoregressive Language Models
A classifier trained only on transformer fine-tuning data detects an invariant memorization signature that transfers to Mamba, RWKV-4, and RecurrentGemma with AUCs of 0.963, 0.972, and 0.936.
-
LiveBench: A Challenging, Contamination-Limited LLM Benchmark
LiveBench is a contamination-limited LLM benchmark with auto-scored challenging tasks from recent sources across math, coding, reasoning and more, where top models score below 70%.
-
OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments
OSWorld provides the first unified real-computer benchmark for open-ended multimodal agent tasks, exposing large performance gaps between humans and state-of-the-art LLM/VLM agents.
-
A Sensitivity-Aware Test Collection for Search Among Personal Information
A new sensitivity-labeled test collection is released from Enron emails with crowdsourced queries, relevance judgments, and LLM extensions for evaluating sensitivity-aware search.
-
Every Act Has Its Price: Compressed Moral Composition in Frontier LLMs
Moral Trolley Arena shows frontier LLMs produce composite moral preferences that are compressed rather than additive functions of calibrated component act strengths across Moral Foundations Theory.
-
Beyond English and Evasion: A Human-Annotated Multi-Domain Benchmark for High-Stakes LLM Safety Evaluation in Chinese
Introduces ChiSafe-PAS, a 1,897-prompt human-annotated Chinese adversarial benchmark for LLM safety with 3-class labels, 9-category obfuscation taxonomy, and domain coverage in self-harm, drugs, fraud, and satire.
-
EvoGM: Learning to Merge LLMs via Evolutionary Generative Optimization
EvoGM uses a dual-generator architecture with cycle-consistent learning on winner-loser pairs from search history to optimize LLM merging coefficients inside a multi-round evolutionary pipeline and reports outperformance over baselines on seen and unseen tasks.
-
VisAnalog: A Diagnostic Suite for Visual Concept Transfer on Natural Images
VisAnalog is a new controlled benchmark showing VLMs substantially underperform humans on visual concept transfer under one- to four-step deterministic transformations, with relation inference as the main failure mode.
-
Towards Camera-Robust 3D Localization: Equation-Anchored Tool-Use for MLLMs
Proposes an equation-anchored tool-use method for MLLMs that writes the pinhole back-projection equation in Chain-of-Thought and substitutes retrieved camera intrinsics and depths to achieve robustness in 3D object detection and visual grounding under rescaled intrinsics.
-
Incantation: Natural Language as the Action Interface for Multi-Entity Video World Models
Incantation is the first video world model to use per-frame natural language conditioning for simultaneous multi-entity control and concept-level cross-entity transfer in interactive video generation.
-
TIDAL: Recovering Temporal Phase for Cloud Block Storage Placement from LLM-Derived Semantics
TIDAL recovers temporal phase signals from LLM-derived semantics of provisioning metadata to enable complementary CVD placement, reducing overload frequency by 79.1% on production traces.
-
Semantic Reranking at Inference Time for Hard Examples in Rhetorical Role Labeling
RISE is an inference-time semantic reranking framework that refines low-confidence predictions in rhetorical role labeling using contrastively learned label representations, delivering an average +9.15 macro-F1 gain on hard examples across eight datasets and seven models.
-
Dynamic Chunking for Diffusion Language Models
DCDM replaces positional blocks with learnable semantic chunks via differentiable Chunking Attention, yielding consistent gains over block and unstructured diffusion baselines up to 1.5B parameters.
-
ClawForge: Generating Executable Interactive Benchmarks for Command-Line Agents
ClawForge is a generator framework that creates reproducible executable benchmarks for command-line agents under state conflict, with ClawForge-Bench showing frontier models reach at most 45.3% strict accuracy and that state inspection drives most performance gaps.
-
BlockVLA: Accelerating Autoregressive VLA via Block Diffusion Finetuning
BlockVLA accelerates autoregressive VLA models by 3.3x using block diffusion finetuning, with faster training convergence and better early performance on long-horizon robotic tasks.
-
Efficient and Adaptive Human Activity Recognition via LLM Backbones
Pretrained LLMs adapted via convolutional projections and LoRA act as efficient frozen backbones for sensor-based human activity recognition, delivering strong data efficiency and cross-dataset transfer.
-
ReCrit: Transition-Aware Reinforcement Learning for Scientific Critic Reasoning
ReCrit frames critic interaction as a correctness-transition problem and uses quadrant-based RL rewards to improve LLM performance on scientific reasoning benchmarks by rewarding corrections and robustness while penalizing sycophancy.
-
V-ABS: Action-Observer Driven Beam Search for Dynamic Visual Reasoning
V-ABS is an action-observer beam search method with entropy-based adaptive weighting and an 80k-sample SFT dataset that delivers 19.7% average gains on visual reasoning tasks for MLLMs.
-
Can Agents Price a Reaction? Evaluating LLMs on Chemical Cost Reasoning
LLM agents reach only 50.6% accuracy on chemical cost estimation within 25% error even with tools, dropping with noise due to parsing, pack selection, and tool-use failures.
-
How Many Iterations to Jailbreak? Dynamic Budget Allocation for Multi-Turn LLM Evaluation
DAPRO provides the first dynamic, theoretically guaranteed way to allocate interaction budgets across test cases for bounding time-to-event in multi-turn LLM evaluations, achieving tighter coverage than static conformal survival methods.
-
AstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical Classification
AstroAlertBench evaluates multimodal LLMs on astronomical classification accuracy, reasoning, and honesty using real ZTF alerts, revealing that high accuracy often diverges from self-assessed reasoning quality.
-
StoryAlign: Evaluating and Training Reward Models for Story Generation
StoryReward, trained on a new 100k story preference dataset, sets state-of-the-art performance on the introduced StoryRMB benchmark for aligning LLM stories with human preferences.
-
When Is the Same Model Not the Same Service? A Measurement Study of Hosted Open-Weight LLM APIs
Hosted open-weight LLM APIs function as time-varying heterogeneous services rather than fixed model artifacts, with demand concentrated, supply-use mismatches, and task-specific routing yielding major cost and throughput gains.
-
LearnPruner: Rethinking Attention-based Token Pruning in Vision Language Models
LearnPruner prunes vision tokens to 5.5% of the original count while retaining about 95% of VLM performance and delivering 3.2 times faster inference by fixing attention sink in encoders and using unbiased middle-layer attention in LLMs.
-
Provably Secure Steganography Based on List Decoding
List decoding enables a provably secure steganography scheme with higher embedding capacity for LLMs via candidate sets and suffix matching.
-
ParetoSlider: Diffusion Models Post-Training for Continuous Reward Control
ParetoSlider conditions diffusion models on continuous preference weights to approximate the full Pareto front, providing dynamic control over multi-objective rewards at inference time.
-
RefAerial: A Benchmark and Approach for Referring Detection in Aerial Images
RefAerial is a new benchmark dataset for text-based object detection in aerial imagery, accompanied by an SCS model that handles scale diversity better than prior ground-image methods.
-
PlayCoder: Making LLM-Generated GUI Code Playable
PlayCoder raises the rate of LLM-generated GUI apps that can be played end-to-end without logic errors from near zero to 20.3% Play@3 by adding repository-aware generation, agent-driven testing, and iterative repair.
-
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.
-
From Experience to Skill: Multi-Agent Generative Engine Optimization via Reusable Strategy Learning
MAGEO is a multi-agent system that distills validated editing patterns into reusable optimization skills for generative engines, outperforming heuristic baselines on visibility and fidelity via a new benchmark and evaluation protocol.
-
On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability
LLM-based dense retrievers generalize better when instruction-tuned but pay a specialization tax when optimized for reasoning; they resist typos and corpus poisoning better than encoder-only baselines yet remain vulnerable to semantic perturbations, with larger models and certain embedding geometry,
-
Beyond Visual Cues: Semantic-Driven Token Filtering and Expert Routing for Anytime Person ReID
STFER uses LVLM-generated identity-consistent semantic text to drive visual token filtering and expert routing for improved any-time person re-identification under clothing changes and modality shifts.
-
Character Beyond Speech: Leveraging Role-Playing Evaluation in Audio Large Language Models via Reinforcement Learning
RoleJudge is a multidimensional evaluation framework for speech-character alignment in audio LLMs, backed by the RoleChat dataset and multi-stage RL training with standard alignment to reduce reward issues.
-
Structural Anchors and Reasoning Fragility:Understanding CoT Robustness in LLM4Code
CoT prompting in LLM4Code shows mixed robustness that depends on model family, task structure, and perturbations destabilizing structural anchors, leading to trajectory deformations like lengthening, branching, and simplification.
-
Online Reasoning Video Object Segmentation
The work introduces the ORVOS task, the ORVOSB benchmark with causal annotations across 210 videos, and a baseline using updated prompts plus a temporal token reservoir.
-
RESP: Reference-guided Sequential Prompting for Visual Glitch Detection in Video Games
RESP uses reference-guided sequential prompting with VLMs to improve frame-level and video-level visual glitch detection in games by establishing per-video baselines.
-
RAG-KT: Cross-platform Explainable Knowledge Tracing with Multi-view Fusion Retrieval Generation
RAG-KT frames cross-platform knowledge tracing as context-constrained LLM inference by building unified multi-source context via Question Group abstractions and retrieving complementary reliable context for grounded predictions and interpretable diagnosis.
-
Pseudo-Unification: Entropy Probing Reveals Divergent Information Patterns in Unified Multimodal Models
Unified multimodal models exhibit pseudo-unification due to modality-asymmetric entropy encoding and pattern-split responses between text and image generation.
-
EgoTL: Egocentric Think-Aloud Chains for Long-Horizon Tasks
EgoTL provides a new egocentric dataset with think-aloud chains and metric labels that benchmarks VLMs on long-horizon tasks and improves their planning, reasoning, and spatial grounding after finetuning.
-
VISOR: Agentic Visual Retrieval-Augmented Generation via Iterative Search and Over-horizon Reasoning
VISOR is a unified agentic VRAG framework with Evidence Space structuring, visual action evaluation/correction, and dynamic sliding-window trajectories trained via GRPO-based RL that achieves SOTA performance on long-horizon visual reasoning benchmarks.
-
Phonemes vs. Projectors: An Investigation of Speech-Language Interfaces for LLM-based ASR
Phoneme-based interfaces match or surpass projector-based ones for LLM ASR, especially in low-resource languages, and a BPE-phoneme hybrid offers additional improvements.
-
SAGE: A Service Agent Graph-guided Evaluation Benchmark
SAGE is a new multi-agent benchmark that formalizes service SOPs as dynamic dialogue graphs to measure LLM agents on logical compliance and path coverage, uncovering an execution gap and empathy resilience across 27 models in 6 scenarios.
-
TaxPraBen: A Scalable Benchmark for Structured Evaluation of LLMs in Chinese Real-World Tax Practice
TaxPraBen is a new benchmark with 14 datasets and a structured evaluation method for measuring LLM performance on Chinese real-world tax tasks and scenarios.
-
DMax: Aggressive Parallel Decoding for dLLMs
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.
-
How Independent are Large Language Models? A Statistical Framework for Auditing Behavioral Entanglement and Reweighting Verifier Ensembles
A new auditing framework reveals widespread behavioral entanglement among LLMs and shows that reweighting ensembles based on measured independence improves verification accuracy by up to 4.5%.
-
ID-Selection: Importance-Diversity Based Visual Token Selection for Efficient LVLM Inference
ID-Selection combines importance scoring with iterative diversity suppression to prune 97.2% of visual tokens in LVLMs while retaining 91.8% performance and cutting FLOPs by over 97% without retraining.