mcshao commited on
Commit
bb040a3
·
verified ·
1 Parent(s): cb1d6af

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -5
README.md CHANGED
@@ -5,7 +5,7 @@ language:
5
  - en
6
  base_model:
7
  - Qwen/Qwen3-Omni-30B-A3B-Instruct
8
- ---
9
 
10
  # LAT-Audio
11
 
@@ -21,7 +21,7 @@ During reasoning, LAT-Audio iteratively incorporates audio evidence through a **
21
  - temporal hallucination (invalid timestamps)
22
  - timestamp drift (progressive misalignment over time)
23
 
24
- ---
25
 
26
  ## Model Description
27
 
@@ -38,7 +38,7 @@ LAT-Audio formulates long-form audio understanding as a structured reasoning pro
38
 
39
  This design enables robust temporal reasoning under long-context settings, where conventional direct modeling approaches often fail.
40
 
41
- ---
42
 
43
  ## Model Variants
44
 
@@ -48,7 +48,7 @@ We provide two model variants:
48
  | **LAT-Audio** | ✅ Yes | LAT-Chronicle | Tool-augmented multi-step reasoning model with global-to-local temporal inference |
49
  | **LAT-Audio-Base** | ❌ No | LAT-Chronicle + in-house | Direct modeling baseline fine-tuned from Qwen3-Omni with more in-house data, offering faster and simpler inference |
50
 
51
- ---
52
 
53
  ## Quick Start
54
 
@@ -58,10 +58,11 @@ pip install -U "huggingface_hub[cli]"
58
  huggingface-cli download mcshao/LAT-Audio --local-dir ./LAT-Audio
59
  huggingface-cli download mcshao/LAT-Audio-Base --local-dir ./LAT-Audio-Base
60
 
 
61
  For detailed inference methods and examples, please refer to the official repository:
62
  👉 https://github.com/alanshaoTT/LAT-Audio-Repo
63
 
64
- ---
65
 
66
  ## Contact
67
 
 
5
  - en
6
  base_model:
7
  - Qwen/Qwen3-Omni-30B-A3B-Instruct
8
+
9
 
10
  # LAT-Audio
11
 
 
21
  - temporal hallucination (invalid timestamps)
22
  - timestamp drift (progressive misalignment over time)
23
 
24
+
25
 
26
  ## Model Description
27
 
 
38
 
39
  This design enables robust temporal reasoning under long-context settings, where conventional direct modeling approaches often fail.
40
 
41
+
42
 
43
  ## Model Variants
44
 
 
48
  | **LAT-Audio** | ✅ Yes | LAT-Chronicle | Tool-augmented multi-step reasoning model with global-to-local temporal inference |
49
  | **LAT-Audio-Base** | ❌ No | LAT-Chronicle + in-house | Direct modeling baseline fine-tuned from Qwen3-Omni with more in-house data, offering faster and simpler inference |
50
 
51
+
52
 
53
  ## Quick Start
54
 
 
58
  huggingface-cli download mcshao/LAT-Audio --local-dir ./LAT-Audio
59
  huggingface-cli download mcshao/LAT-Audio-Base --local-dir ./LAT-Audio-Base
60
 
61
+
62
  For detailed inference methods and examples, please refer to the official repository:
63
  👉 https://github.com/alanshaoTT/LAT-Audio-Repo
64
 
65
+
66
 
67
  ## Contact
68