Signal Use and Emergent Cooperation
arXiv:2506.18920v1 Announce Type: new Abstract: In this work, we investigate how autonomous agents, organized into tribes, learn to use communication…
MuseControlLite: Multifunctional Music Generation with Lightweight Conditioners
arXiv:2506.18729v2 Announce Type: replace-cross Abstract: We propose MuseControlLite, a lightweight mechanism designed to fine-tune text-to-music generation models for precise conditioning…
FAF: A Feature-Adaptive Framework for Few-Shot Time Series Forecasting
arXiv:2506.19567v1 Announce Type: cross Abstract: Multi-task and few-shot time series forecasting tasks are commonly encountered in scenarios such as the…
PrivacyXray: Detecting Privacy Breaches in LLMs through Semantic Consistency and Probability Certainty
arXiv:2506.19563v1 Announce Type: cross Abstract: Large Language Models (LLMs) are widely used in sensitive domains, including healthcare, finance, and legal…
ReDit: Reward Dithering for Improved LLM Policy Optimization
arXiv:2506.18631v2 Announce Type: replace-cross Abstract: DeepSeek-R1 has successfully enhanced Large Language Model (LLM) reasoning capabilities through its rule-based reward system.…
Quantum-Classical Hybrid Quantized Neural Network
arXiv:2506.18240v1 Announce Type: cross Abstract: Here in this work, we present a novel Quadratic Binary Optimization (QBO) model for quantized…
Re-Evaluating Code LLM Benchmarks Under Semantic Mutation
arXiv:2506.17369v1 Announce Type: cross Abstract: In the era of large language models (LLMs), code benchmarks have become an important research…
CUBA: Controlled Untargeted Backdoor Attack against Deep Neural Networks
arXiv:2506.17350v1 Announce Type: cross Abstract: Backdoor attacks have emerged as a critical security threat against deep neural networks in recent…
LightRetriever: A LLM-based Hybrid Retrieval Architecture with 1000x Faster Query Inference
arXiv:2505.12260v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs)-based hybrid retrieval uses LLMs to encode queries and documents into low-dimensional…
SEAL: Scaling to Emphasize Attention for Long-Context Retrieval
arXiv:2501.15225v2 Announce Type: replace-cross Abstract: While many advanced LLMs are designed to handle long sequence data, we can still observe…