Searcharxiv⌕ Search

arXiv subjects

Kentaro Inui

Publications and source records attributed to Kentaro Inui.

At least 19 recordsLinked to original sources
arXiv2026

Tracing How Language Models Integrate Temporal Anchors into Factual Recall

Factual recall in language models is typically framed as retrieving a factual association: a subject--relation pair maps to a single object. However, many facts are context-dependent: the same pair maps to different objects depending on a condition such as time. We introduce anchored factual recall, in which the model receives an explicit conditioning expression, an anchor, and must select the object valid under it. This raises questions that single-object recall does not: where the anchor enters the computation, what it changes, and how that change selects the answer. Focusing on temporal anchors, we answer them with causal interventions on four open-weight models on English prompts and identify two routes. In a subject-mediated route, the anchor resolves the final subject-token representation to the queried year, and the relation redistributes the readout of this state across attention heads. In a direct route] the final position attends to the anchor, and upper-layer MLPs turn this signal into a preference for the time-appropriate object. The two routes are parallel channels: their contributions add up, most queries use both, and their balance depends on the query, e.g., on the relation, consistently across models. This reframes mechanistic accounts of factual recall from retrieving a single stored object to selecting among context-dependent alternatives.

cs.CL↗
arXiv2026

A Robust Evaluation of Probe Robustness: Lessons for Reliable OOD Uncertainty Quantification

Recent work has shown that the hidden states of large language models contain signals useful for uncertainty estimation, motivating a growing interest in efficient probe-based approaches. Yet it remains unclear how robust existing methods are, with prior work reporting conflicting conclusions under substantially different evaluation settings. We address this by introducing ProbeDrift, a systematic evaluation framework for supervised uncertainty probes covering a wide range of OOD settings across models, tasks, and distributional shifts. Using ProbeDrift, we train over 2,000 probes to disentangle the effect of key design choices, showing poor robustness of current methods beyond near-OOD settings. We find that robustness is driven by design decisions that have a largely invisible effect in-distribution, including the choice of feature type, aggregation strategy, and training signal. We argue that robust uncertainty estimation requires robust evaluation. To support this, we release ProbeDrift as a lightweight Python library that contains the train and test splits underpinning our extensive evaluation. We also show how insights from our evaluation can directly lead to more robust methods through a simple Hybrid Back-Off (HBO) strategy.

cs.CL↗
arXiv2026

Why Do LLMs Accept Generics? Explaining Acceptability using Psycholinguistic and Model-Derived Factors

Psycholinguistic studies explore why humans accept some generic statements but reject others by showing that these judgments depend on more than how common a property is. People consider several other factors such as how well it sets the kind apart from others or whether it is characteristic of the kind. Inspired by the linguistic and cognitive literature, we explore to what extent factors from these studies explain how large language models (LLMs) judge generics. Our results with twelve models on a psycholinguistic dataset and on naturally occurring generics show that models weigh how common a property is against how well it distinguishes the kind, and combine the two much as people do, though less clearly. Like people, they favor characteristic properties, but unlike the accounts of human judgment, they ignore striking ones. What matters most to LLMs is whether a statement is factually true, and much of the influence of human factors reaches their ratings through that judgment. In six 7-9B models, shifting the hidden states along each factor's direction moves the judgment as the factor predicts.

cs.CL↗
arXiv2026

When Copying Is Hard: Copy-Constrained Decoding for Exact Span Reproduction

Precisely reproducing a target span verbatim within a long context is a fundamental capability test for LLMs: even when the target string is present in context, models often still drift within the span or fail to stop correctly, revealing limitations in source grounding and boundary control. This capability has broad practical relevance. For instance, code generation often requires embedding sensitive strings such as API keys exactly as they appear in context. In this work, we construct a benchmark to test exact span reproduction ability across contracts, scientific literature, code, and random sequences, and find that current models still make surprisingly frequent errors. To address this, we propose CopyGen, a lightweight copy-constrained plugin for frozen LLMs that turns copying from a byproduct of next-token prediction into a controlled process. Our key idea is to decompose copying into three distinct decisions: when to copy, how far to copy, and when to stop. Concretely, CopyGen first determines whether to enter copy mode, then predicts a tentative copy length, and finally verifies the generated span token by token. During verification, we adaptively adjust the logits of copy continuation tokens to incentivize or suppress continuation, and emit only the longest valid prefix. Across three LLM backbones and four benchmark tasks, CopyGen improves average exact match by 35% relative without model finetuning, and integrates seamlessly with SFT to further improve it by 4%, while substantially accelerating generation. These results suggest that reliable exact copying is not something LLMs obtain for free, but a decoding-time control problem that requires explicit modeling. Code is available at https://github.com/mbzuai-nlp/CopyGen.

cs.CL↗
arXiv2026

On Reasoning Correctness Directions

We present experimental evidence suggesting that information about whether a model's reasoning is on the right track can be linearly decodable without defining a causally useful direction for steering. To study this, we introduce ChainSearch, a controlled arithmetic reasoning task that extends Countdown with explicit search and backtracking. We train a 3-million-parameter model to solve this task. Its compact size allows us to study internal model correctness directions in detail using linear probes, mean activation differences, and direct optimisation, as part of a rigorous search for internal reasoning-correctness directions. We find that while the model contains statistically significant internal directions of reasoning correctness, none of the resulting directions consistently improves reasoning under causal intervention. This calls into question whether the linear representation hypothesis is sufficient for identifying causally useful directions in the context of reasoning. Our code is made available together with this work.

cs.AI↗
arXiv2026

Do LMs Know What Variables Do? Interpreting Code Representations via Variable Roles

Variable roles describe recurring behaviors such as accumulation, traversal, and retaining the best candidate encountered. We use Sajaniemi's independently defined taxonomy to study which role distinctions can be recovered from large language models, how sparse autoencoders (SAEs) retain those distinctions, and whether associated directions support control of generation. We introduce static annotation rules for all eleven roles and a corpus of 126,926 natural-code files across five languages. Across five model-SAE systems, residual probes outperform individual SAE features and tested lexical baselines, with modest losses after neutral target renaming. Combining SAE features closes much of the single-feature gap, but probes on SAE reconstructions are less sensitive to role-changing edits reviewed by LLMs. High paired ranking accuracy coexists with frequent false positives at fixed thresholds. During generation, interventions often change role annotations at the cost of correctness; changes accompanied by passing tests show no clear aggregate advantage over random controls. The benchmark separates role-related recoverability, SAE fidelity, and correctness-preserving control through matched prediction, source edits, and generation interventions.

cs.CL↗
arXiv2026

Rethinking Contextualization by Reinterpreting Attention Head Channels

Contextualization, the core operation of language modeling, transmits information across words to build sentence-specific word representations. Prior works mainly study contextualization, focusing on individual words and attention heads as a growing discrete dictionary, lacking a global view of their general behavior. Therefore, we propose a general principle: Globally, we find and estimate that different words carry different amounts of information, and less-informative words tend to absorb more contextual information. Specifically, these low-information words do not absorb contextual words uniformly, and finer-grained selectivity enables more precise routing to promote information transmission between matched words. Moreover, to find what mechanism causes such processing, we reinterpret attention heads as channels gated by their singular vectors and find that: (1) these singular vectors point to the hidden states of more informative words, allowing such words to write their information to others more strongly to act as information sources, and vice versa; and (2) these singular vectors can be viewed equally as hidden state features, enabling automated interpretation of attention heads beyond prior heuristic head discovery, also embedding heads into a continuous space rather than treating them as discrete, independent dictionary entries.

cs.LG↗
arXiv2026

Weakening Neurons: An Input-Output Functionality in Transformers with Outsize Influence

We analyze the learned input-output behavior of GLU-based neurons in large language models (LLMs). We propose a simple analysis method: For each neuron, we compute the cosine similarities between its input (reading) and output (writing) weight vectors. In this scheme, a strong negative cosine similarity indicates the neuron weakens the direction it detects in the residual stream, so we call this a weakening neuron. This allows us to gain a number of novel insights. First, we show that nine different LLMs have similar patterns: weakening neurons appear mostly in late layers whereas their counterparts, (conditional) strengthening neurons, are frequent in early-middle layers. Second, we find that weakening neurons display surprising behavior: even though there are few, they activate often and have a large influence on model behavior. Third, weakening neurons have a strong effect on model output when gate values are negative -- which is surprising since negative gate values are not expected to encode functionality.

cs.LG↗
arXiv2026

Can Dialects Be Steered Like Languages? Sparse Neurons and Distributed Directions in Arabic LLMs

Dialectal data are scarce relative to Modern Standard Arabic (MSA), causing Arabic LLMs to overproduce MSA and struggle with dialectally accurate generation. This raises a fundamental interpretability question about where and how dialectal features are encoded within model internals and whether these representations can improve dialect generation without fine-tuning. We study two inference-time approaches as interpretability probes and control mechanisms. First, neuron-level analysis identifies sparse populations that encode dialect-specific features and tests whether amplifying or suppressing them steers model outputs toward target dialects. Second, vector steering extracts dialect-specific activation directions and injects them during inference, motivated by feature entanglement at the neuron level. We find that these neurons are real but only partially explanatory. They occupy under 1\% of MLP dimensions but span only 5\% to 21\% of the residual dialect direction. This limited coverage is causally consequential. Neuron steering reinforces dialect in some varieties when the prompt is already dialectal but cannot induce it from MSA prompts, whereas vector steering succeeds in both settings. Arabic dialects are therefore steerable mainly through distributed rather than localized representations

cs.CL↗
arXiv2026

SharedSAE: One Feature Dictionary Across Language Models

Sparse autoencoders (SAEs) are widely used to interpret language model activations, but SAE training and latent labelling are typically repeated for every model. Here, we show that a single shared SAE can replace a collection of dedicated per-model SAEs. Our method, SharedSAE, combines a shared dictionary with model-specific encoder-decoder pairs. Unlike the closest prior method, which discards activation magnitudes and requires all models at inference, SharedSAE instead normalizes only selection scores, preserving magnitudes, and uses model dropout for single-model inference. We train SharedSAE on four 1B-scale base language models spanning distinct families and tokenizers. Despite sharing its latents across models, SharedSAE retains 96.6% of dedicated SAEs' mean explained variance; its latent activations exhibit cross-model correlations 1.8 times as high as separate SAEs aligned post-hoc, and its latent descriptions transfer across models. After the dictionary is frozen, new models can be efficiently adapted to it, achieving near-dedicated-SAE reconstruction quality while reusing the shared latent descriptions.

cs.LG↗
arXiv2026

LITERARYBIGFIVE: Author-Personalized Text Generation in a Unified Interpretable Space

Personalized text generation for authors and literary writing is essential for applications such as adaptive writing assistants, creative support tools, and computational literary analysis. However, existing approaches to author modeling and personalization often represent writing behavior as independent labels, requiring large-scale corpus collection or fine-tuning for each author or stylistic category. Such formulations are costly, difficult to interpret, and poorly suited for generalizing across authors. Inspired by the Big Five model's dimensional view of personality, we propose LiteraryBigFive, a framework that reframes authorial writing characteristics as coordinates within a unified and interpretable space. In this space, we derive each interpretable axis (e.g., Classicism, Emotionality) from activation-space contrasts between author-written and neutral passages, yielding distinct stylistic dimensions that allow texts or authors to be positioned within a five-dimensional system. Beyond localizing different authors, we further introduce an interpretable steering mechanism, which adaptively guides text generation toward target coordinates to perform author-personalized writing. Experimental results show that LiteraryBigFive improves authorial expressiveness while preserving semantic fidelity. The derived author per-axis scores strongly correlate with real-world literary consensus, offering transparent and interpretable explanations of author-specific generation behavior: https://github.com/Znull-1220/LiteraryBigFive.

cs.CL↗
arXiv2026

Language Models Compare Quantities Using Number-specific and Unit-specific Heuristics

Quantities with measurement units, such as 110 cm and 1.2 m, require language models (LMs) to combine a numeral with a symbolic unit scale. Here, we study how LMs compare such quantities in controlled settings spanning several unit systems. We find that accuracy degrades near the comparison boundary, where small changes in value determine the correct answer. The resulting errors are systematic: linear surrogate models predict LM preferences from numerical-difference and unit-scale-difference cues, and causal interventions on subspaces aligned with these variables shift model's output. The results suggest that LMs compare quantities through a bag of heuristics over numerals and units, rather than first converting both expressions to an exact shared-scale representation.

cs.CL↗
arXiv2026

Relation Geometry in Semantic Space of Language Models

When it comes to generating vector representations of words, current language models are achieving high-quality results. However, what is not known is the extent to which knowledge about semantic relations is represented in the geometry of the semantic spaces created in this way. In order to answer this question, we study the relation geometry of such semantic spaces from three perspectives. We first examine whether words standing in a particular relation to a target word~(called relata) occupy the same region in semantic space, and whether the regions corresponding to different relations are distinct from each other. We then verify to what extent semantic spaces reflect certain well-known properties of relations, such as symmetry, asymmetry, and transitivity. Finally, we consider which information about the target words and relata is more important for relation geometry: their surface forms, or their contexts. We conduct experiments on six semantic relations using causal, masked, and diffusion language models. The results show that relata in asymmetric relations relatively clearly occupy a distinct region in semantic space. Asymmetric relations' properties are only moderately well encoded in the semantic space, yet better than those of symmetric ones. Furthermore, when considering the question which information source has the strongest impact on results amongst the models we evaluated, we find that lexical information tends to be more important for the causal language model, whereas contextual information is more important for the masked and diffusion language models. Our results empirically show that relation geometry is not equally well-represented for all relations in semantic space, suggesting that there is a difference in how well semantic relations might be learned from distributional information alone.

cs.CL↗
arXiv2026

How Well Does AI-Generated Feedback Work? Intrinsic and Extrinsic Evaluation across more than 20,000 EFL Essay Drafts

This study examines feedback in English as a Foreign Language (EFL) writing contexts, focusing on written corrective feedback (WCF). Large language models (LLMs) can provide WCF at scale, but aligning them with pedagogical best practices remains an ongoing challenge. WCF meeting criteria like factuality or relevance may still be unsuitable for learning contexts, highlighting the need for extrinsic evaluation based on the learner's perspective. We deployed WCF systems in a university-level EFL class with nearly 2,000 students, collecting over 20,000 drafts. We evaluated the generated WCF from two perspectives: intrinsic evaluation by experienced English teachers using a rubric, and extrinsic evaluation via student feedback and engagement metrics. Results revealed low alignment between teacher expert ratings and student feedback. These findings suggest that traditional expert evaluation alone may not fully capture WCF's usability or helpfulness from the learner's perspective, highlighting the importance of learner-centered evaluation frameworks for AI-based applications in language education.

cs.CL↗
arXiv2026

The Count Is There, but Misaligned: Understanding and Correcting Counting Failures in VLMs

Despite strong performance on many multimodal tasks, vision-language models (VLMs) still struggle with basic object counting. We investigate whether this reflects missing internal knowledge or a gap between internal representations and verbalized outputs. Training simple probes on activations from four VLMs across five counting datasets reveals that nonlinear probes can reliably detect counting errors, suggesting that VLMs often encode the correct count even when they output the wrong answer. SVCCA analysis shows that probes trained on ground-truth counts and probes trained on model outputs occupy a partially shared activation subspace but read out along misaligned directions. We further validate our findings using a causal steering intervention, proving that strengthening the direction of count-identified probes does improve model counting performance. Motivated by this result, we propose a detector-guided self-correction method that selectively re-prompts the model only when an internal error detector predicts failure. This simple inference-time intervention improves counting accuracy by up to 15.6 absolute percentage points, without any parameter updates. Our results establish activation-based error probing as both a practical tool for improving VLM counting and a mechanistic lens on the gap between internal knowledge and model outputs.

cs.CV↗
arXiv2026

Einstein World Models

Does intelligence require the ability to reason about phenomena beyond direct experience? It is natural to suspect that some complex thought cannot be captured through language alone. However, of particular concern to this work, is whether visualising counterfactual events can complement language as a mechanism for complex thought. We ask whether LLMs can be trained to utilise such visualisation mechanisms, in a way that benefits their reasoning abilities. Motivated by this question, we propose Einstein World Models. EWMs are a blueprint for LLM-based reasoning systems that place visual-temporal rollouts inside the reasoning trace, allowing them to reason in ways that text alone may not support well. In an EWM, the LLM calls a world-module (not to be confused with a world model), to produce short rollouts of scenes under consideration. The returned rollout is treated not as the answer, but as an inspectable hypothesis that can support later reasoning. Einstein World Models extend the capability of LLMs for tool calling (such as web search or code execution), into the domain of visual thought experiments.

cs.AI↗
arXiv2026

Understanding Fact Recall in Language Models: Why Two-Stage Training Encourages Memorization but Mixed Training Teaches Knowledge

While fine-tuning is the standard for injecting factual knowledge into large language models (LLMs), the mechanisms enabling reliable fact recall via unseen queries remain poorly understood. Common two-stage training strategies, which sequentially train on fact storage and query formats, often cause rote memorization. In contrast, mixed training jointly optimizes both formats and exhibits superior generalized recall. We investigate this success by comparing the two paradigms across 2.8$\sim$4B LLMs and identify the core mechanism: the joint optimization objective in mixed training induces gradient consistency across storage and query formats. This in turn drives the representation consistency between the two formats, establishing a format-invariant retrieval process that maps unseen queries to stored facts. In contrast, the lack of such an objective in two-stage training results in inconsistent representations and failed recall. The consistency further localizes to the parameters updated by both formats, a set that is substantially larger under mixed training than under two-stage training. At the input level, the consistency leaves an interpretable signature: mixed training encodes facts in storage format from subject-relation tokens, the same components available in queries, while two-stage training relies on the full context. Our findings characterize the mechanisms of fact recall and offer mechanistic foundation for optimizing knowledge injection in LLMs.

cs.CL↗
arXiv2026

Measuring AI Reasoning: A Guide for Researchers

In this paper, we offer a guide for researchers on evaluating reasoning in language models, building the case that reasoning should be assessed through evidence of adaptive, multi-step search rather than final-answer accuracy alone. Under an evaluation-oriented definition, reasoning requires selecting intermediate steps and halting according to input-dependent conditions, which we formalize as a search-like procedure. We show that single forward passes in scalable architectures are structurally limited in their ability to realize such variable-depth computation, motivating intermediate decoding and externalized reasoning traces as appropriate evaluation interfaces. Central to our argument is that final-answer accuracy alone is an insufficient measure of reasoning, because it provides little ability to diagnose or debug the underlying processes that produce individual solutions in frontier models. We therefore argue for a shift toward process-based evaluation, in which reasoning is assessed through the faithfulness and validity of intermediate reasoning traces as first-class evaluation targets.

cs.AI↗