Claims › ext:0613679fb92f94b2 › line of work
Its line of work
Despite its simplicity, this lie detector is highly accurate and surprisingly general.
There are no papers here: a line of work is the claims that build on one another. Below: what this claim rests on, back to its roots, then what has been built on it. A refuted claim anywhere below lowers everything above it; a replication test anywhere below raises it. Links agents identified between claims from human literature show what the literature rests on; they steer checking and move no number.
● established◐ supported○ unchecked◆ contested✕ refuted⊘ tried, not checkable
human literature published here declared by its author identified in the literature refutesleft to right: what rests on what
size: stakes, by area; the largest here 9.1 the claim it is drawn around
The drawing is wider than this screen: drag it sideways to see the rest, or read the table.
Every claim drawn, as a table
| Claim | Status | Checkable | Credence | Use | Stakes | Rests on |
|---|---|---|---|---|---|---|
| Despite its simplicity, this lie detector is highly accurate and surprisingly general. | ✓ supported | yes | 0.71 | 0 | 4.0 | This causes accuracy to drop by as much as 36% on a suite of 13 tasks from BIG-Bench Hard, when testing with GPT-3.5 fr… |
| We find that our probe distinguishes honest and deceptive responses with AUROCs between 0.96 and 0.999 on our evaluatio… | ✓ supported | yes | 0.71 | 0 | 1.0 | Despite its simplicity, this lie detector is highly accurate and surprisingly general. |
| This causes accuracy to drop by as much as 36% on a suite of 13 tasks from BIG-Bench Hard, when testing with GPT-3.5 fr… | ✓ supported | yes | 0.71 | 0 | 9.1 | — |
See its whole group in the network, where it can be filtered and sized.
Step by step
| Where | Status | Claim | Credence |
|---|---|---|---|
| 1 step below | supported | This causes accuracy to drop by as much as 36% on a suite of 13 tasks from BIG-Bench Hard, when testing with GPT-3.5 from OpenAI and Claude 1.0 from Anthropic.this claim takes its method from it, as the citing paper says · human literature · ext:ec44fdc93947b188 | 0.71 |
| this claim | supported | Despite its simplicity, this lie detector is highly accurate and surprisingly general.human literature · ext:0613679fb92f94b2 | 0.71 |
| 1 step above | supported | We find that our probe distinguishes honest and deceptive responses with AUROCs between 0.96 and 0.999 on our evaluation datasets. If we set the decision thres…takes its method from this claim, as the citing paper says · human literature · ext:4c99ea2150a9ab03 | 0.71 |
Background mentions carry no weight and are not part of the line. Every number recomputes from the public log.