File size: 2,185 Bytes
39b7684
51c043a
39b7684
 
51c043a
 
 
 
 
 
 
39b7684
 
51c043a
39b7684
51c043a
 
 
39b7684
51c043a
 
 
 
 
 
 
39b7684
51c043a
39b7684
51c043a
39b7684
51c043a
 
 
 
 
 
 
39b7684
51c043a
 
 
 
 
39b7684
51c043a
39b7684
51c043a
39b7684
51c043a
39b7684
51c043a
 
 
 
 
 
 
39b7684
51c043a
 
 
 
 
39b7684
51c043a
39b7684
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
---
license: apache-2.0
pipeline_tag: text-generation
tags:
  - lfm2
  - data-use
  - provenance
  - usage
  - impact
  - sft
  - lora
---

# lfm2.5-350M-multitask-datause

LoRA SFT of `LiquidAI/LFM2.5-350M` for data-mention provenance attributes
(producer / year / geography / acronym) and usage/impact classification
(data_type / usage_action / impact_label / usage_summary).

## Training
- base model: `LiquidAI/LFM2.5-350M`
- dataset: `rafmacalaba/data-use-sft`
- epochs: 5
- learning rate: 0.0002
- LoRA: r=16 alpha=32 dropout=0.05
- completion-only masking (loss on assistant JSON turn)

## real holdout

Holdout n=14776. Exact string match of each emitted attribute against the gold label.

| attribute | tp | fp | fn | precision | recall | f0.5 | f1 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| producer | 1909 | 968 | 914 | 0.6635 | 0.6762 | 0.6660 | 0.6698 |
| year | 2264 | 520 | 631 | 0.8132 | 0.7820 | 0.8068 | 0.7973 |
| geography | 2632 | 984 | 968 | 0.7279 | 0.7311 | 0.7285 | 0.7295 |
| acronym | 1973 | 456 | 449 | 0.8123 | 0.8146 | 0.8127 | 0.8134 |
| **overall** | 8778 | 2928 | 2962 | 0.7499 | 0.7477 | 0.7494 | 0.7488 |

Usage/impact macro-F1 (per head):
- data_type: 0.7005
- usage_action: 0.6554
- impact_label: 0.5842
- usage_summary: mean_sim=0.5522 grounded_rate=0.6513

Verbatim rate (emitted values that are substrings of the context): 11659/11706 = 0.9960

## synthetic holdout

Holdout n=2305. Exact string match of each emitted attribute against the gold label.

| attribute | tp | fp | fn | precision | recall | f0.5 | f1 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| producer | 231 | 39 | 56 | 0.8556 | 0.8049 | 0.8449 | 0.8294 |
| year | 265 | 50 | 40 | 0.8413 | 0.8689 | 0.8466 | 0.8548 |
| geography | 344 | 63 | 53 | 0.8452 | 0.8665 | 0.8494 | 0.8557 |
| acronym | 145 | 84 | 36 | 0.6332 | 0.8011 | 0.6609 | 0.7073 |
| **overall** | 985 | 236 | 185 | 0.8067 | 0.8419 | 0.8135 | 0.8239 |

Usage/impact macro-F1 (per head):
- data_type: 0.7818
- usage_action: 0.6489
- impact_label: 0.5147
- usage_summary: mean_sim=0.5752 grounded_rate=0.6768

Verbatim rate (emitted values that are substrings of the context): 1216/1221 = 0.9959