“2026-01-19”版本间的差异

来自cslt Wiki
跳转至： 导航、 搜索
 
（13位用户的25个中间修订版本未显示）
第6行： 第6行：
 
|Dong Wang
 
|Dong Wang
 
||
 
||
−
*
+
* Recheck AI textbooks for primary and middle schools.
 
||
 
||
 
*
 
*
第17行： 第17行：
 
|Lantian Li
 
|Lantian Li
 
||
 
||
−
*
+
* Final review of my MLA book (6/10, hold...)
 +
* MoE daily work (busy week)
 +
* My Prof. Title defense
 
||
 
||
 
*
 
*
第28行： 第30行：
 
|Wenqiang Du
 
|Wenqiang Du
 
||
 
||
−
*
+
* Statistics the accounts for companies, our laboratory and AIGE.
 +
* Check AI textbooks
 
||
 
||
 
*
 
*
第39行： 第42行：
 
|Yang Wei
 
|Yang Wei
 
||
 
||
−
*
+
* Analyze MD experiment with 3 annotator labels [https://z1et6d3xtb.feishu.cn/docx/YGHjdXhHIouqZZxZuYmcnOJnnJb?from=from_copylink]
 +
* Develop streaming audio visual speech separation, still with latency
 
||
 
||
 
*
 
*
第76行： 第80行：
 
|Lily
 
|Lily
 
||
 
||
−
*
+
* Participated in reviewing the AI handbooks (primary/middle/high school)
 +
* Compiled course materials for the AI Handbook (Senior Edition)
 
||
 
||
 
*
 
*
第87行： 第92行：
 
|Pengqi Li
 
|Pengqi Li
 
||
 
||
−
*
+
* Writing the Experiments section.
 +
** Focus on hard-to-explain results by formulating hypotheses and conducting deeper analysis.
 
||
 
||
 
*
 
*
第98行： 第104行：
 
|Junming Yuan
 
|Junming Yuan
 
||
 
||
−
*
+
* Aug-MT-HuBERT(failed)
 +
** Continued pre-training for 600K steps. there is still no improvement observed on clean-speech tasks.
 +
*** Aug-MT-HuBERT has more substitution errors in clean ASR adaptation.
 +
* Incorporate a "learn-not-to-listen" mechanism into MT-HuBERT and retrain the backbone. (in progress)
 +
** At 425K steps, PR(PER%): 8.18%, ASR(WER%): 9.32%, SD(DER%): 5.05%
 +
* middle school AI textbook checking(picture, table)
 
||
 
||
 
*
 
*
第109行： 第120行：
 
|Yu Zhang
 
|Yu Zhang
 
||
 
||
−
*
+
* GPU Util: [https://z1et6d3xtb.feishu.cn/wiki/XX4NwX3tJiBDcgkMi0hcFUtInHh]
 +
* LLM
 +
** Rewrite the graph connectivity of Swarm MMLU (the original topology does not align with our expectations).
 +
** Add context segmentation logic to the ECS computation, so that we can trace which preceding node a given text segment originates from.
 
||
 
||
 
*
 
*
第120行： 第134行：
 
|Junhui Chen
 
|Junhui Chen
 
||
 
||
−
*
+
* LLM
 +
** Get swarm crosswords experiment with metric detection working
 +
*** because crosswords has more different role characters than MMLU, so it is necessary
 +
** Check and find some bugs in code of @ZhangYu
 +
** Implement multi-process of LLM instances in experiments
 +
*** p.s.: try multi-threading, fails because a single Python interpreter only initializes one CUDA context, causing resource contention between different instances.
 
||
 
||
 
*
 
*
第130行： 第149行：
 
|-
 
|-
 
|Jiaying Wang
 
|Jiaying Wang
 +
||
 +
* loudness verification 2/3
 +
* spk experiment 3-mix testing, 2-mix & 4-mix training
 
||
 
||
 
*
 
*
 +
||
 +
*
 +
|-
 +
 +
 +
|-
 +
|Xiaoxue Luo
 +
||
 +
* 2-5mix multi_head separation model for Huawei project [https://z1et6d3xtb.feishu.cn/docx/JtLidXiOVoQUedx3h5IcQOwAnFh]
 +
* prepare 2mix speech separation results for ASR test
 
||
 
||
 
*
 
*
第142行： 第174行：
 
|Bochao Hu
 
|Bochao Hu
 
||
 
||
−
*
+
* read some papers
 +
* refactor P2S code, using ms-swift and adding n-best seq to input.
 
||
 
||
 
*
 
*
第153行： 第186行：
 
|Hongcheng Zhang
 
|Hongcheng Zhang
 
||
 
||
−
*
+
*Complete the training of short audio descriptions for AudioSet-SL in WavCaps
 +
**METEOR: 0.3373， ROUGE-L: 0.3121， BLEU-4 : 0.0769， CIDEr : 0.6207
 
||
 
||
 
*
 
*
第164行： 第198行：
 
|Weiman Sun
 
|Weiman Sun
 
||
 
||
−
* Label audioset vedio dataset
+
* Label audioset video dataset
 
* Analyze the representation of MT-LLM inference and review relevant literature
 
* Analyze the representation of MT-LLM inference and review relevant literature
 
||
 
||

2026年1月19日 (一) 11:50的最后版本

People This Week Next Week Task Tracking (DeadLine)
Dong Wang
  • Recheck AI textbooks for primary and middle schools.
Lantian Li
  • Final review of my MLA book (6/10, hold...)
  • MoE daily work (busy week)
  • My Prof. Title defense
Wenqiang Du
  • Statistics the accounts for companies, our laboratory and AIGE.
  • Check AI textbooks
Yang Wei
  • Analyze MD experiment with 3 annotator labels [1]
  • Develop streaming audio visual speech separation, still with latency
Ying Shi
  • Thesis
  • Huawei multi-talker ASR project(Mandarin 2-mix test with FunASR-Nano-800M)
    • ground-truth WER 3.89%
    • separated WER 12.11%
Yue Gu
  • write my Phd thesis
  • the seminar
Lily
  • Participated in reviewing the AI handbooks (primary/middle/high school)
  • Compiled course materials for the AI Handbook (Senior Edition)
Pengqi Li
  • Writing the Experiments section.
    • Focus on hard-to-explain results by formulating hypotheses and conducting deeper analysis.
Junming Yuan
  • Aug-MT-HuBERT(failed)
    • Continued pre-training for 600K steps. there is still no improvement observed on clean-speech tasks.
      • Aug-MT-HuBERT has more substitution errors in clean ASR adaptation.
  • Incorporate a "learn-not-to-listen" mechanism into MT-HuBERT and retrain the backbone. (in progress)
    • At 425K steps, PR(PER%): 8.18%, ASR(WER%): 9.32%, SD(DER%): 5.05%
  • middle school AI textbook checking(picture, table)
Yu Zhang
  • GPU Util: [2]
  • LLM
    • Rewrite the graph connectivity of Swarm MMLU (the original topology does not align with our expectations).
    • Add context segmentation logic to the ECS computation, so that we can trace which preceding node a given text segment originates from.
Junhui Chen
  • LLM
    • Get swarm crosswords experiment with metric detection working
      • because crosswords has more different role characters than MMLU, so it is necessary
    • Check and find some bugs in code of @ZhangYu
    • Implement multi-process of LLM instances in experiments
      • p.s.: try multi-threading, fails because a single Python interpreter only initializes one CUDA context, causing resource contention between different instances.
Jiaying Wang
  • loudness verification 2/3
  • spk experiment 3-mix testing, 2-mix & 4-mix training
Xiaoxue Luo
  • 2-5mix multi_head separation model for Huawei project [3]
  • prepare 2mix speech separation results for ASR test
Bochao Hu
  • read some papers
  • refactor P2S code, using ms-swift and adding n-best seq to input.
Hongcheng Zhang
  • Complete the training of short audio descriptions for AudioSet-SL in WavCaps
    • METEOR: 0.3373， ROUGE-L: 0.3121， BLEU-4 : 0.0769， CIDEr : 0.6207
Weiman Sun
  • Label audioset video dataset
  • Analyze the representation of MT-LLM inference and review relevant literature
取自“http://index.cslt.org/mediawiki/index.php?title=2026-01-19&oldid=43890”