“2026-03-23”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（5位用户的9个中间修订版本未显示）
第19行： 第19行：
 
|Lantian Li
 
|Lantian Li
 
||
 
||
−
*
+
* MoE daily work
 
||
 
||
 
*
 
*
第28行： 第28行：
  
 
|-
 
|-
−
|Wenqiang Du
 
−
||
 
 
|Wenqiang Du
 
|Wenqiang Du
 
||
 
||
 
*AIbabel
 
*AIbabel
 
** Deployment funasr-nano  
 
** Deployment funasr-nano  
−
** find the reason for the low KWS ACC(current thresholding strategy is wrong.）
+
** find the reason for the low KWS ACC(current threshold setting strategy is wrong）
−
**Local deployment the Qwen3-Omni-30B-A3B-Instruct and Testing  Qwen's audio-video understanding capabilities.（90%）
+
**Local deployment the Qwen3 and Testing  Qwen's audio-video understanding capabilities（90%）
 
* Audio separation
 
* Audio separation
−
* add causal convolution (80/100 epoch)
+
** add causal convolution (80/100 epoch)
−
* add slft-attention (50/100 epoch)
+
** add slft-attention (50/100 epoch)
 
* AIGE
 
* AIGE
 
** AI 100 question (60/100)
 
** AI 100 question (60/100)
 
** High school AI couse done（14/14）
 
** High school AI couse done（14/14）
−
||
 
−
*
 
 
||
 
||
 
*
 
*
第52行： 第48行：
 
|Yang Wei
 
|Yang Wei
 
||
 
||
−
*
+
* Audio separation stream inference. (sliding window; in progress)[https://z1et6d3xtb.feishu.cn/docx/MUdRd9v0AoenS8xgRrtcR67qnCe]
 +
* Support MDPI rebuttal.
 
||
 
||
 
*
 
*
第148行： 第145行：
 
|Jiaying Wang
 
|Jiaying Wang
 
||
 
||
−
*
+
* thesis modification
 
||
 
||
 
*
 
*
第172行： 第169行：
 
|Bochao Hu
 
|Bochao Hu
 
||
 
||
−
*
+
* GongAn vts
 +
** problem: stoi is difficult to satisfy and test dataset size they want is too large.
 +
** result: cnvsrc-multi is added to train vts, stoi metric becomes better, but it's not good enough.
 +
** solution: more data is used to train vts(CN-CVS+CN-CVS2-P1+CN-CV3)
 
||
 
||
 
*
 
*
第183行： 第183行：
 
|Hongcheng Zhang
 
|Hongcheng Zhang
 
||
 
||
−
*
+
*reproduce the TAC（Timestamped Audio Captioning）'s Dynamic Acoustic Mixer（30%）
 
||
 
||
 
*
 
*

2026年3月23日 (一) 11:09的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • AIGE book (University version 1/4)
  • Keep on NMI paper proof editing
Lantian Li
  • MoE daily work
Wenqiang Du
  • AIbabel
    • Deployment funasr-nano
    • find the reason for the low KWS ACC(current threshold setting strategy is wrong）
    • Local deployment the Qwen3 and Testing Qwen's audio-video understanding capabilities（90%）
  • Audio separation
    • add causal convolution (80/100 epoch)
    • add slft-attention (50/100 epoch)
  • AIGE
    • AI 100 question (60/100)
    • High school AI couse done（14/14）
Yang Wei
  • Audio separation stream inference. (sliding window; in progress)[1]
  • Support MDPI rebuttal.
Ying Shi
  • Revise my thesis according to Prof. Han's suggestions.
  • Learn some new skills for my future job
Yue Gu
  • write my thesis
Lily
  • Revise thesis
  • Journal paper rebuttal
Pengqi Li
  • Compiled the existing experimental results into a full paper draft.
  • Traveled to Taiyuan to attend the interview and take the required physical examination.
Junming Yuan
  • Preparing the materials for attending ICASSP 2026 (40%)
  • ICASSP Oral PPT (done)
  • MT-HuBERT Journal paper refinement(5/6)
Yu Zhang
  • GPU Util [2]
  • Iterative chain-based LLM experiments:
    • PKS steadily decreases as useful information accumulates (reflecting reduced uncertainty)
    • When no new useful information is introduced, increasing the number of reasoning iterations alone no longer reduces PKS.
Junhui Chen
  • Reorganized the narrative structure and identified additional experiments to include （objectives and design of the Chain structure and Single Node settings）
Jiaying Wang
  • thesis modification
Xiaoxue Luo
  • USS with TDA(transformer decoder-based attractor)
    • write code and train the model(in training)
  • prepare my resume for Huawei internship
Bochao Hu
  • GongAn vts
    • problem: stoi is difficult to satisfy and test dataset size they want is too large.
    • result: cnvsrc-multi is added to train vts, stoi metric becomes better, but it's not good enough.
    • solution: more data is used to train vts(CN-CVS+CN-CVS2-P1+CN-CV3)
Hongcheng Zhang
  • reproduce the TAC（Timestamped Audio Captioning）'s Dynamic Acoustic Mixer（30%）
Weiman Sun
  • read the paper shared by hongcheng
  • fine-tune the mutli-talker llm
Ge Gao
  • convert the MTASS framework to a streaming architecture
  • prepare my graduation project
取自“http://cslt.org/mediawiki/index.php?title=2026-03-23&oldid=44037”