Frequency-aware remasking lifts diffusion language model test-time scaling
MATH-500 gains reach 3.73 points for LLaDA-8B-Instruct and 7.33 points for Dream-7B-Instruct.
Stories per day · 30 days
2 RSS
Relateddiffusion-language-models1distillation1gguf1llm-training1math-5001ollama1quantcheck1quantization1qwen1reasoning-traces1remasking1test-time-scaling1traceforge1
7 stories, newest first · one card per story
Stories tagged #humaneval
Some official Ollama low-bit GGUF tags fail code tasks despite intact weights
Seven of 305 official Ollama tags solved at most six of 164 HumanEval+ tasks.
TraceForge writes student-style reasoning traces for distillation
With the same 5,000 questions and teacher answers, HumanEval+ rises 3.7 points and MBPP+ 7.1.
Reasoning with Continuous Latent Diffusion
Latent Flow Reasoning Models generate math and code solutions by iterative denoising in continuous latent space.
Post-Training Leaves Behavioral Shadows on Unrelated Decisions
Active Taskless Distillation transfers coding and reasoning skills using one unrelated teacher word per prompt.
Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs
Flash-dLLM speeds diffusion LLM inference with I/O-aware KV caching and parallel draft-and-verify decoding.
Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs
Flash-dLLM accelerates diffusion LLM inference with I/O-aware KV caching, up to 11 times faster than Elastic-Cache.