Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -64,7 +64,8 @@ As the number of updates per key (N) increases, LLMs confuse earlier values with
|
|
| 64 |
|
| 65 |
## Cognitive science connection: Proactive Interference (PI)
|
| 66 |
Our test adopts the classic proactive interference paradigm from cognitive science, a foundational method for studying human working memory. PI shows how older, similar information disrupts encoding and retrieval of newer content. Bringing this approach to LLMs allows us to directly measure how interference—not just context length—limits memory and retrieval.
|
| 67 |
-
|
|
|
|
| 68 |
|
| 69 |
See: https://sites.google.com/view/cog4llm
|
| 70 |
|
|
|
|
| 64 |
|
| 65 |
## Cognitive science connection: Proactive Interference (PI)
|
| 66 |
Our test adopts the classic proactive interference paradigm from cognitive science, a foundational method for studying human working memory. PI shows how older, similar information disrupts encoding and retrieval of newer content. Bringing this approach to LLMs allows us to directly measure how interference—not just context length—limits memory and retrieval.
|
| 67 |
+
|
| 68 |
+
- Interestingly, humans are also affected by these three dimensions, but far less than LLMs. Humans consistently outperforming even the latest and largest models on this task.”
|
| 69 |
|
| 70 |
See: https://sites.google.com/view/cog4llm
|
| 71 |
|