Entities · Models
Qwen3-8B
49 articles tagged with this entity.
-
Harness-Aware Self-Evolving: Co-Evolving Model Weights, Harness, and Task Solutions
-
Out-of-Distribution Generalization of Risk Aversion in Language Models
-
HiFA4: Training-Free 4-bit FlashAttention on Ascend HIF4 NPUs for LLM Inference
-
World Feedback for Clinical Agents: Diagnosing RL in FHIR Environments
-
A Mechanistic View of Authority Hierarchy in LLM Sycophancy
-
Fine-Tuning General-Purpose Large Language Models for Agricultural Applications:A Reproducible Framework and Evaluation Protocol Based on Qwen3-8B
-
Discovering Millions of Interpretable Features with Sparse Autoencoders
-
OPERA: Aligning Open-Ended Reasoning via Objective Perplexity-based Reinforcement Learning
-
Unintended Effects of Geographic Conditioning in Large Language Models
-
PolyKV: Heterogeneous Retention and Allocation for KV Cache Compression
-
Vocabulary Dropout for Curriculum Diversity in LLM Co-Evolution
-
Stop When Further Reasoning Won't Help: Attention-State Adaptive Generation in Reasoning Models
-
When to Write and When to Suppress: Route-Specialized Dual Adapters for Memory-Assisted Knowledge Editing
-
Perceive, Interact, Reason: Building Tool-Augmented Visual Agents for Spatial Reasoning
-
Sparrow: Sparse Rollout for Stable and Efficient Long-context RL of Large Language Models
-
SigmaScale: LLM Compression with SVD-based Low-Rank Decomposition and Learned Scaling Matrices
-
Job Searcher
-
Closing the Loop on Latent Reasoning via Test-Time Reconstruction
-
Emotion-Aware Image Generation from Korean Diary Text via LLM-based Prompt Translation and LoRA Fine-Tuning
-
HarnessForge: Joint Harness and Policy Evolution for Adaptive Agent Systems
-
From "Weak" Signals to Strong Models: Preference Delta Aggregation with LoRA Merging
-
LinguIUTics at PsyDefDetect: Iterative Imbalance-Aware Fine-tuning of Qwen3-8B for Psychological Defense Mechanism Classification
-
Extreme Low-Bit Inference in Reasoning Models: Failure Modes and Targeted Recovery
-
DFlare: Scaling Up Draft Capacity for Block Diffusion Speculative Decoding
-
Cost-Aware Diffusion Draft Trees for Speculative Decoding
-
ProtStructQA: A Denotation Threshold in Protein Structural Reasoning
-
Skill or Skip? Learning Selective Skill Invocation in Agentic Tasks via Dual-Granularity Preference Learning
-
unix-ctf: Procedural Environments for Unix-Competence Reinforcement Learning
-
On-Policy Replay for Continual Supervised Fine-Tuning
-
Revisiting the Effectiveness of LLM Pruning for Test-Time Scaling
-
Hurwitz Quaternion Multiplicative Quantization for KV Cache Compression
-
Better Accuracies, Worse Reasoning: A Step-Level Audit of Medical Chain-of-Thought Distillation
-
Prompt Codebooks: Discrete Compositional Optimization for Language Model Instruction Refinement
-
Playing with Words, Improving with Rewards: Training Language Models for Creative Association
-
Restoring the Sweet Spot: Pass-Rate Weighted Self-Distillation for LLM Reasoning
-
ZipRL: Adaptive Multi-Turn Context Compression with Hindsight Response Replay
-
ClinConsensus: A Physician-Calibrated Benchmark for Evaluating Clinical Rubric Coverage in Chinese Medical LLMs
-
The Shape of Overthinking: Backtracking Bursts in Long Reasoning Traces
-
AERIC: Anticipatory Hidden-State Monitoring for Implicit Harmful Dialogue
-
UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents
-
STREAM: A Data-Centric Framework for Mining High-Value Task-Oriented Dialogues from Streaming Media
-
Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals
-
ECHO: Terminal Agents Learn World Models for Free
-
A general tensor-structured compression scheme for efficient large language models
-
Geo-Expert: Towards Expert-Level Geological Reasoning via Parameter-Efficient Fine-Tuning
-
Beyond Literal Translation: Evaluating Cultural Effectiveness in Social Media UGC
-
Towards a Universal Causal Reasoner
-
IntroLM: Introspective Language Models via Prefilling-Time Self-Evaluation
-
FASQ: Flexible Accelerated Subspace Quantization for Calibration-Free LLM Compression