Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

One Head Away from Language Confusion: Single Attention Heads Across Model Families

Jiwon Jeong, Anjelo Jann Laroza, Chaewoon Yoon, Arjun Pillai +3 more

Published

Oct 18, 2026

Citations

0

Trust level

Low

Usefulness score

0/100 (Low)

Extraction confidence

25% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Oct 18, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Should you rely on this paper?

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.

Best use

Background context only

Use if you need

Background context only.

What to verify

Validate the evaluation procedure and quality controls in the full paper before operational use.

Main weakness

This paper looks adjacent to evaluation work, but not like a strong protocol reference.

Human feedback signal
Not explicit
Not explicit in abstract metadata
Evaluation signal
Weak or implicit
Validate from full paper
Usefulness for eval research
0/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Large language models sometimes reply in an unexpected language, a failure known as language confusion. We ablate individual attention heads in 12 model checkpoints spanning seven model series, using FLORES-200 continuations in five languages and Language Confusion Benchmark (LCB) chat replies. In ten models, zeroing one head causes substantially more departures from the expected language than zeroing any other head in its layer; its LCB pass-rate drop persists, though reduced, under mean ablation. Seven heads move 8.0 to 51.5 percent of all FLORES continuations out of the prompt's language, versus at most 1.4 percent for other same-layer heads; the other three mainly affect explicitly requested languages. These seven heads matter less in base models given plain text. In two models, replacing the values a head reads at the requested-language name with another language's mean makes replies fail and shifts some into that language, unlike a random change of equal size. Replacing a head's output with another language's mean redirects 84 to 99 percent of scored non-English monolingual LCB replies into that language in five models. Yet the requested language's mean improves crosslingual replies beyond every control head only in the two Gemma models, whose baselines are low. Code and results are available in the public repository at https://github.com/christian-hoang-04/anonymous-release.

What we could verify

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.

Human Feedback Types

missing

None explicit

No explicit feedback protocol extracted.

"Large language models sometimes reply in an unexpected language, a failure known as language confusion."

Evaluation Modes

missing

None explicit

Validate eval design from full paper text.

"Large language models sometimes reply in an unexpected language, a failure known as language confusion."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Large language models sometimes reply in an unexpected language, a failure known as language confusion."

Benchmarks / Datasets

partial

DROP

Useful for quick benchmark comparison.

"In ten models, zeroing one head causes substantially more departures from the expected language than zeroing any other head in its layer; its LCB pass-rate drop persists, though reduced, under mean ablation."

Reported Metrics

missing

Not extracted

No metric anchors detected.

"Large language models sometimes reply in an unexpected language, a failure known as language confusion."

Benchmarks and datasets

DROP

Reported metrics

No metric terms were extracted from the available abstract.

Human feedback details
Uses human feedback
No
Feedback types
None
Rater population
Not reported
Expertise required
Coding
Evaluation details
Evaluation modes
None
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Large language models sometimes reply in an unexpected language, a failure known as language confusion.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Large language models sometimes reply in an unexpected language, a failure known as language confusion.
  • We ablate individual attention heads in 12 model checkpoints spanning seven model series, using FLORES-200 continuations in five languages and Language Confusion Benchmark (LCB) chat replies.
  • In ten models, zeroing one head causes substantially more departures from the expected language than zeroing any other head in its layer; its LCB pass-rate drop persists, though reduced, under mean ablation.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Contribution summary

  • We ablate individual attention heads in 12 model checkpoints spanning seven model series, using FLORES-200 continuations in five languages and Language Confusion Benchmark (LCB) chat replies.

Why it matters for eval

  • We ablate individual attention heads in 12 model checkpoints spanning seven model series, using FLORES-200 continuations in five languages and Language Confusion Benchmark (LCB) chat replies.

Researcher checklist

  • Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • 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

    Detected: DROP

  • Metric reporting is present

    No metric terms extracted.