IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchies
Large Language Models (LLMs) often fail to maintain instruction hierarchies (IH) when processing multi-source inputs with varying role-level…
Why am I seeing this Ranked on recency
It is here mostly because it is new. Freshness is the largest single contributor to its score, which means nothing about the story except that it is recent.
Nothing much is corroborating this yet. On a quiet day in developer that is enough to reach the top of the section — recency, not significance, is doing the work here.
Link-outLink-out, because it scores 0.42, below the 0.50 bar for a write-up. Link-out means we point at the publisher and say nothing of our own.
| Factor | Weight | Score | Contribution | Where it came from |
|---|---|---|---|---|
| Corroboration | 0.35 | 0.39 | +0.135 32% | 1 independent org on the story. Tier-3 aggregators never corroborate — they can show something is circulating, never that it is true. |
| Source trust | 0.25 | 0.47 | +0.117 28% | arXiv is the highest-trust source on this story and is first-party — the organisation announcing its own news. Trust is taken from the best source, not averaged. |
| Pickup rate | 0.20 | 0.00 | +0.000 0% | One counted organisation, so there is no spread to measure — nothing has picked this up to set a rate. |
| Freshnessleads | 0.20 | 0.85 | +0.171 40% | Halves every 10 hours from the newest item on the story. This is the only factor that rewards a story for nothing more than being recent. |
Corroboration counts distinct organisations, once each, and only from tiers 1 and 2. Freshness halves every 10 hours, so this ranking is a snapshot and will differ at the next build.
What happened
Large Language Models (LLMs) often fail to maintain instruction hierarchies (IH) when processing multi-source inputs with varying role-level priorities, paradoxically adhering to lower-priority directives during conflicts. While existing defenses mitigate this issue, they are largely restricted to single-turn scenarios and require expensive fine-tuning. In this paper, we formalize this failure mode in multi-turn contexts via a Jensen-Shannon Divergence (JSD) framework, uncovering a pervasive role-influence inversion phenomenon where subordinate inputs override superior roles. To rectify this without training, we propose IHDec (Instruction Hierarchy-steered Decoding).
How this story arrived
Ordered by when each source was first observed, which is what the velocity figure is computed from. Publishers backdate; observed order does not.
- 01 Arxivfirst-party first seen IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchi
Overclock clusters coverage from independent sources and grades it automatically. The figures above are computed, not editorial. This page summarises and links to reporting by the outlets named — follow the links for the original work.