Tian-Fantasea/test123

模型相关论文可查阅:

7B

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration
TiFRe: Text-guided Video Frame Reduction for Efficient Video Multi-modal Large Language Models
Searching Meta Reasoning Skeleton to Guide LLM Reasoning
Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2
OSUM-Pangu: An Open-Source Multidimension Speech Understanding Foundation Model Built upon OpenPangu on Ascend NPUs
Structured Episodic Event Memory
Quasar: Quantized Self-Speculative Acceleration for Rapid Inference via Memory-Efficient Verification
CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns
Beyond Static Question Banks: Dynamic Knowledge Expansion via LLM-Automated Graph Construction and Adaptive Generation
Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking
EvoOpt-LLM: Evolving industrial optimization models with large language models
Accelerating OpenPangu Inference on NPU via Speculative Decoding
V-CAGE: Context-Aware Generation and Verification for Scalable Long-Horizon Embodied Tasks
Scheduling LLM Inference with Uncertainty-Aware Output Length Predictions
Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Prompt
A-IO: Adaptive Inference Orchestration for Memory-Bound NPUs
Multi-source Heterogeneous Public Opinion Analysis via Collaborative Reasoning and Adaptive Fusion: A Systematically Integrated Approach
Characterize LSM-tree Compaction Performance via On-Device LLM Inference
Traceable Cross-Source RAG for Chinese Tibetan Medicine Question Answering
A Comprehensive Evaluation of LLM Reasoning: From Single-Model to Multi-Agent Paradigms
Structured Self-Consistency: A Multi-Task Evaluation of LLMs on VirtualHome
ArkEval: Benchmarking and Evaluating Automated CodeRepair for ArkTS
FMBench: Adaptive Large Language Model Output Formatting
DScheLLM: Enabling Dynamic Scheduling through a Fine-Tuned Dual-System Large language Model
Holmes: An Evidence-Grounded LLM Agent for Auditable DDoS Investigation in Cloud Networks
Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows
How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks

1B

Post-Training Quantization of OpenPangu Models for Efficient Deployment on Atlas A2
VisionPangu: A Compact and Fine-Grained Multimodal Assistant with 1.7B Parameters
Neuromem: A Granular Decomposition of the Streaming Lifecycle in External Memory for LLMs
CSSBench: Evaluating the Safety of Lightweight LLMs against Chinese-Specific Adversarial Patterns
A-IO: Adaptive Inference Orchestration for Memory-Bound NPUs
AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning
Combining Adam and its Inverse Counterpart to Enhance Generalization of Deep Learning Optimizers
Characterize LSM-tree Compaction Performance via On-Device LLM Inference
PanguMotion: Continuous Driving Motion Forecasting with Pangu Transformers
FMBench: Adaptive Large Language Model Output Formatting
Cochain: Balancing Insufficient and Excessive Collaboration in LLM Agent Workflows
How Few-shot Demonstrations Affect Prompt-based Defenses Against LLM Jailbreak Attacks
Adaptive Confidence Gating in Multi-Agent Collaboration for Efficient and Optimized Code Generation

7B-V1.1

SpatialText: A Pure-Text Cognitive Benchmark for Spatial Understanding in Large Language Models
Efficient Reasoning with Balanced Thinking
OSUM-Pangu: An Open-Source Multidimension Speech Understanding Foundation Model Built upon OpenPangu on Ascend NPUs
EvoOpt-LLM: Evolving industrial optimization models with large language models
Accelerating OpenPangu Inference on NPU via Speculative Decoding
CAMERA: Multi-Matrix Joint Compression for MoE Models via Micro-Expert Redundancy Analysis
TypePro: Boosting LLM-Based Type Inference via Inter-Procedural Slicing
UniPruning: Unifying Local Metric and Global Feedback for Scalable Sparse LLMs
Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression
DScheLLM: Enabling Dynamic Scheduling through a Fine-Tuned Dual-System Large language Model
VAR-MATH: Probing True Mathematical Reasoning in LLMS via Symbolic Multi-Instance Benchmarks

1B-V1.1

Silent Inconsistency in Data-Parallel Full Fine-Tuning: Diagnosing Worker-Level Optimization Misalignment
AdapShot: Adaptive Many-Shot In-Context Learning with Semantic-Aware KV Cache Reuse
Probe and Skip: Self-Predictive Token Skipping for Efficient Long-Context LLM Inference

详细清单

Searching Meta Reasoning Skeleton to Guide LLM Reasoning