SeaWolf-AI commited on
Commit
2f0de8a
·
verified ·
1 Parent(s): a9b46e5

Minimize card: remove method and third-party component details (trade secret)

Browse files
Files changed (1) hide show
  1. README.md +3 -22
README.md CHANGED
@@ -2,36 +2,17 @@
2
  license: apache-2.0
3
  tags:
4
  - retrieval
5
- - reasoning-retrieval
6
  - bright
7
- - information-retrieval
8
  language:
9
  - en
10
  ---
11
 
12
  # Darwin-BRIGHT-Retriever
13
 
14
- A reasoning-augmented retrieval **system** for the BRIGHT reasoning-intensive retrieval benchmark.
15
 
16
- This is **not a single model**. It is a two-stage retrieval pipeline:
17
-
18
- 1. **Query reasoning augmentation** with our Darwin reasoning model: for each query, the system first reasons about what an ideal answer passage must contain (key concepts, theorems, terms, entities), and appends that reasoning to the query.
19
- 2. **Dense retrieval** with the open embedder **Qwen3-Embedding-4B** over the augmented query.
20
-
21
- The contribution here is the Darwin reasoning-augmentation stage. The embedder is a public, third-party model used as-is; it is named for full transparency and is not claimed as ours.
22
 
23
  ## Evaluation
24
 
25
- Measured with the MTEB evaluation framework on the **BRIGHT (v1.1)** benchmark (nDCG@10, 20 tasks), under the same protocol as the public leaderboard.
26
-
27
- Baselines reported alongside for context:
28
- - BM25 (public baseline): reproduced on the same tasks.
29
- - Qwen3-Embedding-4B alone (public embedder): reference.
30
-
31
- > Evaluation in progress. Final aggregate score is published only after all tasks complete and are measured end to end. No partial or estimated numbers are reported.
32
-
33
- ## Method notes
34
-
35
- - Query-only augmentation; documents are embedded unchanged.
36
- - Darwin generates a short expansion per query (2-3 sentences), then the augmented query is encoded.
37
- - Deterministic, reproducible pipeline over the released BRIGHT public queries and corpora.
 
2
  license: apache-2.0
3
  tags:
4
  - retrieval
 
5
  - bright
 
6
  language:
7
  - en
8
  ---
9
 
10
  # Darwin-BRIGHT-Retriever
11
 
12
+ FINAL-Bench in-house retrieval system for the **BRIGHT** benchmark (reasoning-intensive information retrieval).
13
 
14
+ Internally developed. Method not disclosed.
 
 
 
 
 
15
 
16
  ## Evaluation
17
 
18
+ Measured with the MTEB evaluation framework on **BRIGHT (v1.1)**, nDCG@10, under the public leaderboard protocol. Scores are published only after the full evaluation completes and is measured end to end.