Rethinking Data Selection at Scale: Random Selection is Almost All You Need
Paper
• 2410.09335 • Published
• 16
LLM Training
-
-
From Generalist to Specialist: Adapting Vision Language Models via Task-Specific Visual Instruction Tuning
Paper • 2410.06456 • Published • 37 -
Emergent properties with repeated examples
Paper • 2410.07041 • Published • 8 -
Personalized Visual Instruction Tuning
Paper • 2410.07113 • Published • 70 -
LLM Self-Correction with DeCRIM: Decompose, Critique, and Refine for Enhanced Following of Instructions with Multiple Constraints
Paper • 2410.06458 • Published • 8 -
What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective
Paper • 2410.23743 • Published • 64 -
Constraint Back-translation Improves Complex Instruction Following of Large Language Models
Paper • 2410.24175 • Published • 18 -
GlotCC: An Open Broad-Coverage CommonCrawl Corpus and Pipeline for Minority Languages
Paper • 2410.23825 • Published • 4 -
Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
Paper • 2410.21333 • Published • 12 -
Fictitious Synthetic Data Can Improve LLM Factuality via Prerequisite Learning
Paper • 2410.19290 • Published • 10 -
AutoTrain: No-code training for state-of-the-art models
Paper • 2410.15735 • Published • 59 -
Stronger Models are NOT Stronger Teachers for Instruction Tuning
Paper • 2411.07133 • Published • 38 -
IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization
Paper • 2411.06208 • Published • 21 -
LLM2CLIP: Powerful Language Model Unlock Richer Visual Representation
Paper • 2411.04997 • Published • 39 -
Language Models are Hidden Reasoners: Unlocking Latent Reasoning Capabilities via Self-Rewarding
Paper • 2411.04282 • Published • 37 -
Large Language Models Can Self-Improve in Long-context Reasoning
Paper • 2411.08147 • Published • 65 -
Cut Your Losses in Large-Vocabulary Language Models
Paper • 2411.09009 • Published • 49 -
Search, Verify and Feedback: Towards Next Generation Post-training Paradigm of Foundation Models via Verifier Engineering
Paper • 2411.11504 • Published • 24 -
RedPajama: an Open Dataset for Training Large Language Models
Paper • 2411.12372 • Published • 57 -
SageAttention2 Technical Report: Accurate 4 Bit Attention for Plug-and-play Inference Acceleration
Paper • 2411.10958 • Published • 57 -
Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization
Paper • 2411.10442 • Published • 87 -
VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection
Paper • 2411.14794 • Published • 13 -
TÜLU 3: Pushing Frontiers in Open Language Model Post-Training
Paper • 2411.15124 • Published • 67 -
VisionZip: Longer is Better but Not Necessary in Vision Language Models
Paper • 2412.04467 • Published • 117 -
Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Models
Paper • 2412.02980 • Published • 15 -
MALT: Improving Reasoning with Multi-Agent LLM Training
Paper • 2412.01928 • Published • 45 -
Free Process Rewards without Process Labels
Paper • 2412.01981 • Published • 34 -
On Domain-Specific Post-Training for Multimodal Large Language Models
Paper • 2411.19930 • Published • 31 -
Reverse Thinking Makes LLMs Stronger Reasoners
Paper • 2411.19865 • Published • 23 -
Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
Paper • 2412.05271 • Published • 160 -
MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale
Paper • 2412.05237 • Published • 46 -
Fully Open Source Moxin-7B Technical Report
Paper • 2412.06845 • Published • 11 -
InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions
Paper • 2412.09596 • Published • 97 -
Euclid: Supercharging Multimodal LLMs with Synthetic High-Fidelity Visual Descriptions
Paper • 2412.08737 • Published • 54 -
Lyra: An Efficient and Speech-Centric Framework for Omni-Cognition
Paper • 2412.09501 • Published • 48 -
VisionArena: 230K Real World User-VLM Conversations with Preference Labels
Paper • 2412.08687 • Published • 13 -
Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference
Paper • 2412.13663 • Published • 161 -
No More Adam: Learning Rate Scaling at Initialization is All You Need
Paper • 2412.11768 • Published • 43 -
RobustFT: Robust Supervised Fine-tuning for Large Language Models under Noisy Response
Paper • 2412.14922 • Published • 88 -
An Empirical Study of Autoregressive Pre-training from Videos
Paper • 2501.05453 • Published • 41 -
REINFORCE++: A Simple and Efficient Approach for Aligning Large Language Models
Paper • 2501.03262 • Published • 104 -
Optimizing Large Language Model Training Using FP4 Quantization
Paper • 2501.17116 • Published • 36 -
GuardReasoner: Towards Reasoning-based LLM Safeguards
Paper • 2501.18492 • Published • 88 -
Scaling Pre-training to One Hundred Billion Data for Vision Language Models
Paper • 2502.07617 • Published • 29 -
How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?
Paper • 2502.14502 • Published • 91 -
Stable-SPAM: How to Train in 4-Bit More Stably than 16-Bit Adam
Paper • 2502.17055 • Published • 20 -
An Embarrassingly Simple Defense Against LLM Abliteration Attacks
Paper • 2505.19056 • Published • 6 -
Orthogonal Finetuning Made Scalable
Paper • 2506.19847 • Published • 11 -
Skywork-SWE: Unveiling Data Scaling Laws for Software Engineering in LLMs
Paper • 2506.19290 • Published • 53 -
Scaling RL to Long Videos
Paper • 2507.07966 • Published • 160 -
Step-Audio 2 Technical Report
Paper • 2507.16632 • Published • 74 -
Towards a Unified View of Large Language Model Post-Training
Paper • 2509.04419 • Published • 76 -
Set Block Decoding is a Language Model Inference Accelerator
Paper • 2509.04185 • Published • 54 -
GTAlign: Game-Theoretic Alignment of LLM Assistants for Mutual Welfare
Paper • 2510.08872 • Published • 4