AINeutralarXiv – CS AI · Jun 237/10
🧠Researchers identify specific attention heads in multilingual language models responsible for language switching errors, revealing that instruction tuning reorganizes these circuits to concentrate language identity signals in early layers. The study demonstrates that language selection operates through a distributed but hierarchical mechanism, with compensation patterns following predictable feedforward cascades rather than global diffusion.
AINeutralarXiv – CS AI · Jun 236/10
🧠Researchers introduce TF-RefusalBench, a multilingual benchmark measuring over-alignment in large language models used for criminal law tasks in Swiss courts. The study demonstrates that safety guardrails designed to prevent harmful outputs inadvertently compromise legitimate legal work by refusing to process content describing violent crimes, and proposes abliteration as an effective mitigation technique.
AINeutralarXiv – CS AI · Jun 116/10
🧠Researchers introduce Neural FOXP2, a technique that identifies and steers language-specific neurons in large language models to shift their default behavior from English to other languages like Hindi or Spanish. The method uses sparse autoencoders and spectral analysis to isolate a compact set of control circuits governing language preference, enabling safer, more targeted manipulation of multilingual model behavior.
AINeutralarXiv – CS AI · Jun 106/10
🧠Researchers at arXiv analyzed how large language models introduce distinctive emotional signatures when translating literary works, finding that LLM translations preserve author's voice less effectively than human translations. Post-editing partially corrects these emotional distortions, but MT systems consistently exhibit model-specific emotional fingerprints that deviate from human translation norms.
AINeutralarXiv – CS AI · Jun 26/10
🧠Researchers discovered that large language models fail to refuse harmful requests in low-resource languages not because they lack the underlying safety representations, but because they cannot properly calibrate their safety decisions across languages. A recalibration approach using minimal target-language examples substantially improves refusal rates, suggesting safety alignment failures stem from decision calibration rather than representation gaps.
🧠 Llama
AINeutralarXiv – CS AI · Jun 16/10
🧠Researchers demonstrate that multilingual large language models encode shared confidence features that transfer across languages without retraining. A lightweight linear probe trained on English can predict answer correctness in unseen languages with zero-shot generalization, suggesting confidence estimation mechanisms are language-universal in LLMs.
AIBearisharXiv – CS AI · Apr 136/10
🧠Researchers evaluated how well frontier LLMs like GPT-4o and Gemini interpret story morals across 14 language-culture pairs, finding that while models generate semantically similar outputs to humans, they lack cultural diversity and concentrate on universally shared values rather than culturally-specific moral interpretations.
🧠 GPT-4🧠 Gemini
