Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2603.26829 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917364782596096 |
|---|---|
| author | Oh, Nathaniel Attie, Paul |
| author_facet | Oh, Nathaniel Attie, Paul |
| contents | Language models detect false premises when asked directly but absorb them under conversational pressure, producing authoritative professional output built on errors they already identified. This failure - order-gap hallucination - is invisible to output inspection because the error migrates into the activation space of the safety circuit, suppressed but not erased. We introduce Squish and Release (S&R), an activation-patching architecture with two components: a fixed detector body (layers 24-31, the localized safety evaluation circuit) and a swappable detector core (an activation vector controlling perception direction). A safety core shifts the model from compliance toward detection; an absorb core reverses it. We evaluate on OLMo-2 7B using the Order-Gap Benchmark - 500 chains across 500 domains, all manually graded. Key findings: cascade collapse is near-total (99.8% compliance at O5); the detector body is binary and localized (layers 24-31 shift 93.6%, layers 0-23 contribute zero, p<10^-189); a synthetically engineered core releases 76.6% of collapsed chains; detection is the more stable attractor (83% restore vs 58% suppress); and epistemic specificity is confirmed (false-premise core releases 45.4%, true-premise core releases 0.0%). The contribution is the framework - body/core architecture, benchmark, and core engineering methodology - which is model-agnostic by design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_26829 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Squish and Release: Exposing Hidden Hallucinations by Making Them Surface as Safety Signals Oh, Nathaniel Attie, Paul Machine Learning Artificial Intelligence Language models detect false premises when asked directly but absorb them under conversational pressure, producing authoritative professional output built on errors they already identified. This failure - order-gap hallucination - is invisible to output inspection because the error migrates into the activation space of the safety circuit, suppressed but not erased. We introduce Squish and Release (S&R), an activation-patching architecture with two components: a fixed detector body (layers 24-31, the localized safety evaluation circuit) and a swappable detector core (an activation vector controlling perception direction). A safety core shifts the model from compliance toward detection; an absorb core reverses it. We evaluate on OLMo-2 7B using the Order-Gap Benchmark - 500 chains across 500 domains, all manually graded. Key findings: cascade collapse is near-total (99.8% compliance at O5); the detector body is binary and localized (layers 24-31 shift 93.6%, layers 0-23 contribute zero, p<10^-189); a synthetically engineered core releases 76.6% of collapsed chains; detection is the more stable attractor (83% restore vs 58% suppress); and epistemic specificity is confirmed (false-premise core releases 45.4%, true-premise core releases 0.0%). The contribution is the framework - body/core architecture, benchmark, and core engineering methodology - which is model-agnostic by design. |
| title | Squish and Release: Exposing Hidden Hallucinations by Making Them Surface as Safety Signals |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2603.26829 |