CCTest · Blog
GigaChat Audio: A Time-Aware Open Audio LLM for Long Recordings
Speech & Audio
cctest.ai
Speech & Audio

GigaChat Audio: A Time-Aware Open Audio LLM for Long Recordings

This paper targets one of the hardest problems in long-audio understanding: temporal grounding. The model can answer questions with explicit timestamps over inputs as long as 120 minutes. Its key idea is to interleave periodic time markers with continuous audio tokens and train with large-scale synthetic supervision.

Read more
CCTest · Blog
DiFA Reframes Diffusion Inference as Forward-Aligned State Estimation
Diffusion Models
cctest.ai
Diffusion Models

DiFA Reframes Diffusion Inference as Forward-Aligned State Estimation

DiFA is a training-free inference framework for diffusion models that treats intermediate predictions as correlated observations rather than disposable integration outputs. By building a forward-aligned temporal consensus, it aims to improve stability and preserve fine details during sampling.

Read more