Results for “language-models”

Deduplicating Training Data Makes Language Models Better

VILA: On Pre-training for Visual Language Models

LoRA: Low-Rank Adaptation of Large Language Models

Multimodal Few-Shot Learning with Frozen Language Models

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

NLVR2: A Visual Reasoning Benchmark for Natural Language

GLIP: Grounded Language-Image Pre-training

Large Language and Vision Assistant. Enables visual instruction tuning and image-based conversations. Combines CLIP vision encoder with Vicuna/LLaMA language models. Supports multi-turn image chat, visual question answering, and instruction following. Use for vision-language chatbots or image understanding tasks. Best for conversational image analysis.

CapsFusion: Rethinking Image-Text Data at Scale

Text-to-speech and speech-to-text using fal.ai audio models

LIMA: Less Is More for Alignment

Generate image captions, answer visual questions, and perform image-text retrieval using BLIP-2's Q-Former architecture with frozen vision encoders and LLMs.

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.

Vision-language pre-training framework bridging frozen image encoders and LLMs. Use when you need image captioning, visual question answering, image-text retrieval, or multimodal chat with state-of-the-art zero-shot performance.

Mental models and decision frameworks for building autonomous agents that continuously learn, explore, and accumulate skills in open-ended environments without human supervision

Domain-Driven Design methodology using bounded contexts, ubiquitous language, and strategic/tactical patterns

Text-to-speech and speech-to-text using fal.ai audio models

Vision-language chat: VQA, captioning, image dialogue.

Enables visual instruction tuning and image-based conversations using open-source vision-language models. Supports multi-turn image chat, visual question answering, and image understanding tasks.

Write idiomatic Kotlin 1.9+ with null safety, sealed classes, data classes, extension functions, delegates, collections, inline/reified generics, and expression syntax. Use for Kotlin language constructs, Java-to-Kotlin migration, lazy delegates, data classes, backing properties, or inline/reified generics; keep language constructs in scope even inside response models, and defer coroutine configuration, Android Context/framework, Retrofit/OkHttp client setup, and other library-specific recipes.

Generate natural speech from text using ElevenLabs voice AI, supporting 70+ languages, multiple models, and various output formats.

Run interpretability experiments on neural network internals using nnsight, with optional NDIF remote execution for massive models.

Generates structured VGL JSON for Bria FIBO models, giving deterministic control over objects, lighting, camera, composition, and style instead of natural language prompts.

Train large language models (7B-70B+) with RLHF using PPO, GRPO, DPO, and other algorithms, accelerated by Ray and vLLM for distributed multi-GPU setups.

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.

Provides guidance for training and analyzing Sparse Autoencoders (SAEs) using SAELens to decompose neural network activations into interpretable features. Use when discovering interpretable features, analyzing superposition, or studying monosemantic representations in language models.

Scans a codebase for security vulnerabilities using CodeQL's interprocedural data flow and taint tracking analysis, with support for multiple languages, scan modes, and data extension models.