RESEARCH MAP · 研究脉络
Follow the questions behind the papers.
A publication list is chronological; research is not. This page groups my work by the recurring questions it tries to answer.
Identify where time, memory, or generalization is actually lost before designing the method.
Prefer improvements that remain visible in measurable end-to-end behavior.
Pair ideas with code, datasets, benchmarks, or project pages whenever possible.
DIFFUSION LANGUAGE MODELS
How can parallel generation become genuinely faster?
Diffusion LLMs expose parallelism, but useful acceleration depends on token confidence, denoising dynamics, verification cost, and the way reasoning is represented.
SlowFast Sampling
Alternates exploratory and accelerated stages using certainty, convergence, and positional principles; the homepage reports a 15.63× speedup on LLaDA.
Thinking inside the Mask
Places reasoning directly into the mask-denoising process through in-place prompting.
Self-Speculative Decoding
Uses internal self-drafting and parallel verification to accelerate diffusion language-model inference.
AUDIO & MULTIMODAL LLMS
What information deserves to stay in memory?
Long audio contexts make memory allocation a semantic decision: compression must preserve acoustically and linguistically critical evidence rather than treating every token or head equally.
AudioKV
Prioritizes audio-critical attention heads through semantic-acoustic alignment and Spectral Score Smoothing; the homepage reports 60% lower memory overhead.
AudioMarathon
A benchmark for evaluating long-context audio understanding together with efficiency.
EMBODIED MEDICAL INTELLIGENCE
How can models generalize from perception to physical guidance?
Robotic ultrasonography couples visual representation, anatomy-aware generalization, and real-time action guidance under strong safety and data constraints.
UltraHiT
A hierarchical Transformer architecture for generalizable internal-carotid-artery robotic ultrasonography.
VA-Adapter
A parameter-efficient vision-action adapter for real-time echocardiography probe guidance.
Research has a route as well as a result.
Continue to the journey timeline