HFEPX · Eval paper review
Oscar Gilg, Pierre Beckmann, Daniel Paleka, Patrick Butlin
Published
May 13, 2026
Citations
0
Trust level
Low
Usefulness score
40/100 (Low)
Extraction confidence
45% (Low)
Derived from extracted protocol signals and abstract evidence.
Rater population
Not reported
Signals refreshed
May 18, 2026
This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.
Use this as background context only. Do not make protocol decisions from this page alone.
Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.
Best use
Background context only
Use if you need
Background context only.
What to verify
Read the full paper before copying any benchmark, metric, or protocol choices.
Main weakness
The available metadata is too thin to trust this as a primary source.
Treat as adjacent context, not a core eval-method reference.
If you are doing eval pipeline work, start here
Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour. But models can also adopt different personas which have radically different preferences. How is this implemented internally? Does each persona run on its own preference machinery, or is something shared underneath? We train linear probes on residual-stream activations of Gemma-3-27B and Qwen-3.5-122B to predict revealed pairwise task choices, and identify a genuine preference vector: it tracks the model's preferences as they shift across a range of prompts and situations, and on Gemma-3-27B steering along it causally controls pairwise choice. This preference representation is largely shared across personas: a probe trained on the helpful assistant predicts and steers the choices of qualitatively different personas, including an evil persona whose preferences anti-correlate with those of the Assistant.
These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.
Pairwise Preference
Directly usable for protocol triage.
"Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour."
None explicit
Validate eval design from full paper text.
"Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour."
Not reported
No explicit QC controls found.
"Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour."
Not extracted
No benchmark anchors detected.
"Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour."
Not extracted
No metric anchors detected.
"Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour."
No benchmark or dataset names were extracted from the available abstract.
No metric terms were extracted from the available abstract.
Large language models (LLMs) can be said to have preferences: they reliably pick certain tasks and outputs over others, and preferences shaped by post-training and system prompts appear to shape much of their behaviour.
Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.
Human feedback protocol is explicit
Detected: Pairwise Preference
Evaluation mode is explicit
No clear evaluation mode extracted.
Quality control reporting appears
No calibration/adjudication/IAA control explicitly detected.
Benchmark or dataset anchors are present
No benchmark/dataset anchor extracted from abstract.
Metric reporting is present
No metric terms extracted.
