North Carolina Exploring Cultural Heritage Online

What pass@k Measures When Attempts Share State: Count Runs, Not Attempts

Search trees, planners and self-correcting agents make the attempts of one run share a random state, yet their results are often scored with a pass@k estimator built for independent draws, which public harnesses apply to whatever attempts they receive. Prior work separates the pr...

What pass@k Measures When Attempts Share State: Count Runs, Not Attempts

Search trees, planners and self-correcting agents make the attempts of one run share a random state, yet their results are often scored with a pass@k estimator built for independent draws, which public harnesses apply to whatever attempts they receive. Prior work separates the pr...

Selecting GPTQ Damping from the Calibration Set by Nested-Ridge Leave-One-Out

GPTQ-style post-training quantization adds λI to the calibration Hessian before inverting it, with λ a fixed fraction of the mean diagonal (one percent in the GPTQ authors' code, five percent in a maintained implementation) that ignores how many calibration rows there are. Small...

Auditing the Transferability of a Published Quantization Comparison: The Verdict Is a Function of the Slice

Practitioners choose between released 4-bit conversions of a model from published comparison sentences such as "AWQ consistently outperforms ... GPTQ ... (7B-70B)", although the files they serve, the serving stack and the way quality is measured all differ from what the sentence...

Where 4-Bit Weight Quantization Breaks In-Context Interference Resolution: Layout, Operating Point and Format on a Paired-Associate Battery

Controlled evidence on 4-bit weight quantization and in-context interference is proactive-only, limited to one 4-bit format and behavioural. In 1.5–8B open-weight models we cross proactive and retroactive layouts, with a recency-counterbalanced control, with interference dose, li...

Input Calibration and a Specificity Audit of Auditory-Scene-Analysis Cue Scores in Music Foundation Models

Do music foundation models encode auditory-scene-analysis (ASA) cues, the grouping principles behind source perception? A high cue score alone cannot answer this: the ordering may be inherited from the model's input transform or, for timing, arise from onset positions alone. We i...

Some Speech Codecs Hide Voice Pathology from a Fixed Detector: Directional Fidelity Readouts for Remote Voice Assessment

This manuscript is a non-peer-reviewed preprint.

Inflation-Aware Quantization: Restoring INT4 Reasoning Token-Cost Parity with Distillation

This manuscript is a non-peer-reviewed preprint.

Damping Is a Ridge Penalty: Calibration-Only Selection for Post-Training Quantization under Scarce Calibration

This manuscript is a non-peer-reviewed preprint.

Auditing the Transferability of a Published Quantization Comparison: The Verdict Is a Function of the Slice

This manuscript is a non-peer-reviewed preprint.

Damping Is a Ridge Penalty: Calibration-Only Selection for Post-Training Quantization under Scarce Calibration

This manuscript is a non-peer-reviewed preprint.

Adaptation, Not Priming: A Repulsive In-Context Aftereffect Carried by a Sparse Set of Attention Heads

In-context learning is assumed to assimilate: repeated exposure pulls the next prediction toward the repeated pattern, as induction and copy mechanisms attest. We report the opposite sign for graded scalar attributes, and establish three things about it. The phenomenon. Across fi...

The Verdict Is a Function of the Slice: Auditing a Published Quantization Superiority Claim

A published comparison is read as deployment guidance, and what that reader needs is that its verdict holds on the files they are about to download. Two moves are supposed to make such a reading trustworthy: fix what will be read before the data exist, and reproduce it on a secon...

Adaptation, Not Priming: A Repulsive In-Context Aftereffect Carried by a Sparse Set of Attention Heads

In-context learning is assumed to assimilate: repeated exposure pulls the next prediction toward the repeated pattern, as induction and copy mechanisms attest. We report the opposite sign for graded scalar attributes, and establish three things about it. The phenomenon. Across fi...

Reproduced Numbers, Unlicensed Claims: Preregistered Acceptance Contracts for Post-Training Compression

Reproducing a published table does not settle whether the sentence it supports was ever specified tightly enough to be accepted or rejected as written. We study that gap through the acceptance contract---the six fields a quality-preservation claim must fix before it is testable (...