Update README.md
Browse files
README.md
CHANGED
|
@@ -44,16 +44,7 @@ computation — none require retrieving information from an in-context document.
|
|
| 44 |
|
| 45 |
This dataset accompanies **LOCOS** (Logit-Contribution Scoring). A retrieval-head
|
| 46 |
detector is only meaningful if ablating the heads it identifies degrades
|
| 47 |
-
*contextual retrieval* while leaving *non-retrieval* abilities intact.
|
| 48 |
-
that gap with the **Dissociation Score** at ablation depth $k$:
|
| 49 |
-
|
| 50 |
-
$$\mathrm{DS}(k) = \Delta R(k) - \Delta P(k)$$
|
| 51 |
-
|
| 52 |
-
where $\Delta R(k) = (R_0 - R(k))/R_0$ is the relative drop in NoLiMa ROUGE-L
|
| 53 |
-
(retrieval), and $\Delta P(k) = (P_0 - P(k))/P_0$ is the relative drop in
|
| 54 |
-
aggregate parametric accuracy (mean over the three sources below). $R_0, P_0$ are
|
| 55 |
-
the unablated baselines. A high DS means the heads are retrieval-specific:
|
| 56 |
-
ablating them hurts retrieval far more than parametric recall or arithmetic.
|
| 57 |
|
| 58 |
## Composition
|
| 59 |
|
|
|
|
| 44 |
|
| 45 |
This dataset accompanies **LOCOS** (Logit-Contribution Scoring). A retrieval-head
|
| 46 |
detector is only meaningful if ablating the heads it identifies degrades
|
| 47 |
+
*contextual retrieval* while leaving *non-retrieval* abilities intact.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
## Composition
|
| 50 |
|