News
Batch-Max: Higher LLM Throughput using Larger Batch Sizes and KV Cache Compression
arXiv:2412.05693v3 Announce Type: replace Abstract: Several works have developed eviction policies to remove key-value (KV)...
Aya Vision: Advancing the Frontier of Multilingual Multimodality
arXiv:2505.08751v1 Announce Type: new Abstract: Building multimodal language models is fundamentally challenging: it requires aligning...
AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists
arXiv:2506.08140v1 Announce Type: cross Abstract: Despite long-standing efforts in accelerating scientific discovery with AI, building...
AutoMixer: Checkpoint Artifacts as Automatic Data Mixers
arXiv:2506.21910v1 Announce Type: new Abstract: In language model training, it is desirable to equip models...
Automatically assessing oral narratives of Afrikaans and isiXhosa children
arXiv:2507.13205v1 Announce Type: new Abstract: Developing narrative and comprehension skills in early childhood is critical...
Attention Is Not Always the Answer: Optimizing Voice Activity Detection with Simple Feature Fusion
arXiv:2506.01365v1 Announce Type: cross Abstract: Voice Activity Detection (VAD) plays a key role in speech...
AsyncSwitch: Asynchronous Text-Speech Adaptation for Code-Switched ASR
arXiv:2506.14190v1 Announce Type: new Abstract: Developing code-switched ASR systems is challenging due to language ambiguity...
AssistedDS: Benchmarking How External Domain Knowledge Assists LLMs in Automated Data Science
arXiv:2506.13992v1 Announce Type: cross Abstract: Large language models (LLMs) have advanced the automation of data...
Apple Researchers Reveal Structural Failures in Large Reasoning Models Using Puzzle-Based Evaluation
Artificial intelligence has undergone a significant transition from basic language models to advanced models that...
Apple makes major AI advance with image generation technology rivaling DALL-E and Midjourney
Apple researchers develop STARFlow, a breakthrough AI image generation system that challenges diffusion models used...
Apple and Duke Researchers Present a Reinforcement Learning Approach That Enables LLMs to Provide Intermediate Answers, Enhancing Speed and Accuracy
Long CoT reasoning improves large language models’ performance on complex tasks but comes with drawbacks...
Anthropic study: Leading AI models show up to 96% blackmail rate against executives
Anthropic research reveals AI models from OpenAI, Google, Meta and others chose blackmail, corporate espionage...
