Graded Entity-Familiarity Readouts in Language Models: Polish Adaptation, Cross-Language Robustness, and Refusal Steering

• URL: http://arxiv.org/abs/2607.13568v1
• Date: Wed, 15 Jul 2026 08:06:59 GMT
• Title: Graded Entity-Familiarity Readouts in Language Models: Polish Adaptation, Cross-Language Robustness, and Refusal Steering
• Abstract summary: Familiarity-probe scores separate real from fabricated entities in every family.<n>In a paired experiment on two families, probes retain 96-101% of within-language AUROC when the Polish question stem is replaced with an English one.<n> familiarity probes are competitive among pre-generation abstention gates.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Can a language model estimate its familiarity with an entity before generating an answer? We study activations at the final prompt token in twelve instruction-tuned models from the Bielik, PLLuM, Gemma-4, and Qwen3 families, using a new dataset of 1,440 Polish entities spanning four domains and ten Wikipedia-pageview deciles, plus fabricated controls. Familiarity-probe scores separate real from fabricated entities in every family; in the Polish-adapted Bielik and PLLuM families they additionally track entity popularity (model-mean Spearman $ρ$ 0.28-0.57, versus at most 0.11 in Gemma-4 and Qwen3), a pattern more strongly associated with Polish adaptation than with parameter count in this model sample. In a paired experiment on two families, probes retain 96-101% of within-language AUROC when the Polish question stem is replaced with an English one around unchanged entity names, showing robustness to prompt language in this setting. In Gemma-4-12B, the only model that natively refuses, adding a one-dimensional familiarity direction at a single layer moves refusal rates monotonically in both directions (0.24 to 1.00 on well-known entities; 0.73 to 0.00 on unknown ones). Finally, a calibrated familiarity probe is competitive among pre-generation abstention gates, although post-generation detectors better predict behavioral error on average. These results support a graded pre-generation entity-familiarity readout, and a separation between representational familiarity and the policy that converts it into abstention.

Related papers

• Beyond the Name: Demographic Leakage in De-Identified Résumés and Evaluation Artifacts in LLM Bias Audits [4.736836688901646]
  We investigate whether eliminating language fields resolves leakage across nine open-weight models and 620 résumés.<n>We isolate unstructured prose across five ethnocultural conditions and three cue-salience tiers.<n>Target-group recovery averages 0.757 overall and saturates at 1.000 under high salience.
  arXiv  Detail & Related papers  (2026-09-15T01:48:44Z)
• REGARD: Regional Affective Differences in Large Language Models [32.505127447635864]
  We study what drives affective framing differences across LLMs on post-Soviet entities using target-directed Valence-Arousal-Dominance profiling.<n>Post-hoc Ward-linkage clustering of all 19 models by affective and response-behavior profiles yields three behavioral clusters.<n>These findings show that VAD profiling captures emotional intensity, a dimension of affective framing that is largely invisible to conventional sentiment-based evaluation.
  arXiv  Detail & Related papers  (2026-07-22T20:47:57Z)
• DiaLLM: An Investigation into the Robustness-Generation Gap in English Dialect Adaptation [20.845447775728545]
  Large language models increasingly emphunderstand dialectal English, yet still emphproduce only standard, US-leaning English.<n>We introduce textbfDiaLLM, which continually pretrains three open-weight language model families on the International Corpus of English.
  arXiv  Detail & Related papers  (2026-07-08T17:24:27Z)
• Evaluating Large Language Models for Hausa and Fongbe Machine Translation: Benchmarks, Failures, and Metric Reliability [12.074798555934855]
  We investigate the translation quality of current large language models (LLMs) for English-to-Hausa and English-to-Fongbe.<n>We evaluate whether standard automatic metrics reliably reflect human judgment for these low-resource languages.
  arXiv  Detail & Related papers  (2026-06-20T23:23:54Z)
• Benchmarking Linguistic Adaptation in Comparable-Sized LLMs: A Study of Llama-3.1-8B, Mistral-7B-v0.1, and Qwen3-8B on Romanized Nepali [0.0]
  Romanized Nepali, the Nepali language written in the Latin alphabet, is the dominant medium for informal digital communication in Nepal.<n>This study presents a systematic benchmarking of linguistic adaptation across three comparable-sized open-weight models.
  arXiv  Detail & Related papers  (2026-03-25T07:02:51Z)
• Alignment Backfire: Language-Dependent Reversal of Safety Interventions Across 16 Languages in LLM Multi-Agent Systems [0.0]
  In perpetrator treatment, offenders articulate remorse yet behavioral change does not follow.<n>We show that alignment interventions produce a structurally analogous phenomenon: surface safety that masks or generates collective pathology and internal dissociation.<n>These findings reframe alignment as a behavioral intervention subject to risk homeostasis and iatrogenesis.
  arXiv  Detail & Related papers  (2026-03-05T07:46:59Z)
• Cross-Lingual Stability of LLM Judges Under Controlled Generation: Evidence from Finno-Ugric Languages [0.22009842278462158]
  Cross-lingual evaluation of large language models (LLMs) typically conflates two sources of variance: genuine model performance differences and measurement instability.<n>We investigate evaluation reliability by holding generation conditions constant while varying target language.<n>Our findings suggest that zero-shot judge transfer is unreliable for discourse-level assessment in morphologically rich languages.
  arXiv  Detail & Related papers  (2026-02-02T16:27:32Z)
• Beyond the Final Layer: Intermediate Representations for Better Multilingual Calibration in Large Language Models [50.34755385896279]
  Confidence calibration is crucial for the reliable deployment of Large Language Models (LLMs)<n>We conduct the first large-scale, systematic studies of multilingual calibration across six model families and over 100 languages.<n>We find that non-English languages suffer from systematically worse calibration.
  arXiv  Detail & Related papers  (2025-10-03T16:07:15Z)
• Emulated Disalignment: Safety Alignment for Large Language Models May Backfire! [65.06450319194454]
  Large language models (LLMs) undergo safety alignment to ensure safe conversations with humans. This paper introduces a training-free attack method capable of reversing safety alignment. We name this method emulated disalignment (ED) because sampling from this contrastive distribution provably emulates the result of fine-tuning to minimize a safety reward.
  arXiv  Detail & Related papers  (2024-02-19T18:16:51Z)
• mFACE: Multilingual Summarization with Factual Consistency Evaluation [79.60172087719356]
  Abstractive summarization has enjoyed renewed interest in recent years, thanks to pre-trained language models and the availability of large-scale datasets. Despite promising results, current models still suffer from generating factually inconsistent summaries. We leverage factual consistency evaluation models to improve multilingual summarization.
  arXiv  Detail & Related papers  (2022-12-20T19:52:41Z)
• OneAligner: Zero-shot Cross-lingual Transfer with One Rich-Resource Language Pair for Low-Resource Sentence Retrieval [91.76575626229824]
  We present OneAligner, an alignment model specially designed for sentence retrieval tasks. When trained with all language pairs of a large-scale parallel multilingual corpus (OPUS-100), this model achieves the state-of-the-art result. We conclude through empirical results and analyses that the performance of the sentence alignment task depends mostly on the monolingual and parallel data size.
  arXiv  Detail & Related papers  (2022-05-17T19:52:42Z)
• Few-shot Learning with Multilingual Language Models [66.49496434282564]
  We train multilingual autoregressive language models on a balanced corpus covering a diverse set of languages. Our largest model sets new state of the art in few-shot learning in more than 20 representative languages. We present a detailed analysis of where the model succeeds and fails, showing in particular that it enables cross-lingual in-context learning.
  arXiv  Detail & Related papers  (2021-12-20T16:52:35Z)
• AmericasNLI: Evaluating Zero-shot Natural Language Understanding of Pretrained Multilingual Models in Truly Low-resource Languages [75.08199398141744]
  We present AmericasNLI, an extension of XNLI (Conneau et al.), to 10 indigenous languages of the Americas. We conduct experiments with XLM-R, testing multiple zero-shot and translation-based approaches. We find that XLM-R's zero-shot performance is poor for all 10 languages, with an average performance of 38.62%.
  arXiv  Detail & Related papers  (2021-04-18T05:32:28Z)
• Harnessing Multilinguality in Unsupervised Machine Translation for Rare Languages [48.28540903568198]
  We show that multilinguality is critical to making unsupervised systems practical for low-resource settings. We present a single model for 5 low-resource languages (Gujarati, Kazakh, Nepali, Sinhala, and Turkish) to and from English directions. We outperform all current state-of-the-art unsupervised baselines for these languages, achieving gains of up to 14.4 BLEU.
  arXiv  Detail & Related papers  (2020-09-23T15:07:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.