Improving the Diversity of LLM Outputs without a Trade-off
cs.CL, cs.AI, cs.LG
Submitted: 2026-09-27
Updated: 2026-09-27
Terminology
Sources
- SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Language GANs Falling Short
- Modifying Large Language Model Post-Training for Diverse Creative Writing
- High Quality Rather than High Model Probability: Minimum Bayes Risk Decoding with Neural Metrics
- Stochastic Beams and Where to Find Them: The Gumbel-Top-k Trick for Sampling Sequences Without Replacement
- Free Lunch for Pass@$k$? Low Cost Diverse Sampling for Diffusion Language Models
- No Prompt Left Behind: Exploiting Zero-Variance Prompts in LLM Reinforcement Learning via Entropy-Guided Advantage Shaping
- How Diversely Can Language Models Solve Problems? Exploring the Algorithmic Diversity of Model-Generated Code
- SetPO: Set-Level Policy Optimization for Diversity-Preserving LLM Reasoning
- QuasiMoTTo: Quasi-Monte Carlo Test-Time Scaling
- Jointly Reinforcing Diversity and Quality in Language Model Generations
- Beyond Mode Collapse: Distribution Matching for Diverse Reasoning
- Quasi-random Multi-Sample Inference for Large Language Models
- Qwen2.5 Technical Report
- G2: Guided Generation for Enhanced Output Diversity in LLMs
- Word Tour: One-dimensional Word Embeddings via the Traveling Salesman Problem
- Incremental Sampling Without Replacement for Sequence Models
- Semantic-guided Diverse Decoding for Large Language Model
- Asking a Language Model for Diverse Responses
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering