Named after the hundred-eyed watchman of Greek myth, Argus watches the education landscape: spotting new opportunities, pressure-testing the ventures we're building, and tracing every read back to the real-world signals behind it.
The evidence library: the raw signals the pipeline is watching across the education ecosystem. Every idea is built from these.
arXiv:2608.21747v1 Announce Type: cross Abstract: LLM-based coding agents generate complete software systems from high-level descriptions, yet little is known about how the format of architecture specifications affects the quality of generated code or whether this effect depends on model capability. We present a controlled experiment comparing five informationally equivalent specification formats (informal prose, Mermaid diagrams with constraints and ADRs, OpenAPI, C4/Structurizr DSL, and TypeScript interface contracts with ArchUnit-style rules) across six models from three vendor families (Anthropic Claude, OpenAI GPT, Google Gemini). Across 90 multi-turn agent trials, specification format shows a strong format x model interaction. On the strongest models (Sonnet 4.6, GPT-5), format barely matters (quality spread 0.17-0.92). On weaker models, format produces spreads of 0.83-2.42 points, with code-proximate formats (OpenAPI, TypeScript contracts) recovering most of the capability gap.
arXiv:2608.21713v1 Announce Type: cross Abstract: Reasoning-augmented text-to-image models such as GoT-R1 emit an explicit textual plan - object names, attributes, and bounding boxes - before generating image tokens. When such a model fails a compositional prompt, is the plan wrong, or is the plan right and the decoder unfaithful? Because the plan is machine-readable it can be edited before decoding, which makes the two separable. We first validate the ruler. Swapping the two bounding boxes inside the model's own chain demonstrably flips the generated layout: detector-based accuracy falls 0.75 -> 0.48 (p<1e-3), while a widely used VQA-based spatial metric rises. A five-rater human study agrees with the detector on 81% of items and with the VQA judge on 57%. All spatial results therefore use geometric scoring. Under sound measurement the decoder is a faithful executor: 94% of generated layouts realize the planned relation, and object-box binding survives reordering of the plan's object
arXiv:2608.21702v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) grounds LLM generation on retrieved documents, but the standard terminal retrieval stage--dense-vector similarity, optionally followed by reranking--often returns documents that share keywords with the query without containing the needed information, a failure mode that grows with the knowledge base. We trace it to a conceptual gap: similarity captures only associational relations, whereas the documents that matter are linked to the query causally. We model the terminal retrieval stage with a causal graph grounded in Reichenbach's common cause principle: the keywords shared by the query and a retrieved document form a latent common cause A, and the document's residual keywords form a latent set B linking the document to the ideal output. Since a retrieved document is a collider (A -> d <- B), retrieval itself opens an associational path between the query and B, which licenses a training-free, attenti
arXiv:2608.21664v1 Announce Type: cross Abstract: Safe deployment of increasingly capable models will likely come to rely on latent-space monitoring as a complement to behavioral evaluations, especially when evaluation-aware models exhibit scheming or deception. However, if models can also control their own activations, deception could extend into the latent space itself. With this in mind, we introduce the Activation Controllability Benchmark to quantify the extent to which models can modulate their residual stream via natural-language instruction. Across model families and capability levels, we find that most LLMs can control the direction and magnitude of their residual stream activations with some degree of temporal resolution, though performance varies considerably across models. In simple tasks, this level of control can evade activation-based monitoring methods (including linear probes, natural language autoencoders, activation oracles, and the Jacobian lens), albeit imperfectly
arXiv:2608.21601v1 Announce Type: cross Abstract: Benchmarks for scientific artificial intelligence are mostly written to be scored: multiple-choice questions, curated agent tasks with reference solutions, or simulators with a known generative structure. Real scientific requests arrive differently. They are underspecified, they carry attachments, and lack ground truth. We report K-Bench 01, an evaluation built from first-turn requests sampled from live user traffic on K-Dense Web and run end to end by nine frontier models in identical sandboxes, yielding 1,602 completed agent runs. Three blinded language-model judges scored every run against an eight-dimension rubric. On a rubric whose 8-anchor instructs judges that a domain scientist would accept the work with minor edits, no model clears the line under all three judges. gpt-5.6-sol has the highest pooled mean, 8.04, but its 95% interval [7.80, 8.23] spans the threshold, and two of the three judges rank claude-opus-5 first instead. We
arXiv:2608.21541v1 Announce Type: cross Abstract: KV-cache compression is often justified by attention maps with a few large weights. This is incomplete: large weights may not contain most of the mass, omitted values can cancel, and preserving the attention output may not preserve the task. We separate these questions. Global score gaps -- not threshold counts -- determine how many tokens are needed to retain a target mass. For a realized row, the weighted sum of omitted values is the exact missing statistic. A controlled retrieval--aggregation model explains when truncation helps and when it hurts. These results motivate CertKV, a training-free compressor that reserves one tail-summary slot per head and allocates the rest by value dispersion. Under matched budgets, CertKV is top-two in seven of nine LongBench-v2 settings, remains in the leading compressed tier on 128K RULER, and realizes a ten-fold cache budget in a packed Llama prototype. Compressibility depends on the mass, values,
arXiv:2608.21529v1 Announce Type: cross Abstract: Timely and accurate assessment of property damage is critical following natural disasters. Traditional on-site inspections are labor-intensive, costly, and often pose safety risks. Advances in satellite imagery and vision-language models (VLMs) enable scalable remote damage assessment; however, integrating VLMs into large-scale Earth observation pipelines presents challenges in computational efficiency, data organization, and information retrieval. To address these challenges, we present DamageScope, a retrieval-augmented framework that combines satellite imagery with Vision-Language Models (VLMs) and Large Language Models (LLMs) to automate property damage analysis. Built on a Retrieval-Augmented Generation (RAG) framework, DamageScope extracts structured visual representations from satellite imagery to support interactive natural language queries for damage assessment. To address scalability, we introduce a novel multi-vector embeddin
arXiv:2608.23564v1 Announce Type: new Abstract: Modern software systems accumulate technical debt over decades of development, which makes migration expensive and largely manual. As coding agents become increasingly capable at bug fixing, can they autonomously perform such migrations? Existing benchmarks cannot answer this question because they evaluate only behavioural correctness, not whether the migration actually occurred. This leads an easy hack: agents copy the original implementation to make tests pass. We call this Blindness. To address this problem, we introduce SWE Refactor Bench, a benchmark comprising 20 whole-repository migrations, covering 4 kinds of technical debt. A three-stage evaluation protocol measures both migration completeness and behavioural correctness. (1) Migration Audit verifies that the migration occurred. (2) Behavioural Tests measure correctness with a fixed test suite. (3) Agentic Verification uses 6 independent coding agents to generate targeted tests f
arXiv:2608.23551v1 Announce Type: new Abstract: Recent advances in continuous diffusion and flow-based language models (LMs) have achieved performance competitive with discrete LMs. However, existing continuous frameworks still rely on decoders supervised with cross entropy (CE) because the flow trajectories are not guaranteed to terminate at valid token embeddings. Motivated by this limitation, we introduce \textbf{ConvergeFlow}, an embedding-space flow-based LM, which constrains the data predictor to the convex hull of token embeddings and trains it solely with the mean squared error objective induced by flow matching. Under suitable regularity conditions, we prove that the resulting flow converges to valid token embeddings despite errors in the data predictor, enabling direct token prediction without a CE-supervised decoder. We further develop three sampling mechanisms for controlling the trade-off between the generative perplexity and entropy. Experiments on OpenWebText demonstrate
arXiv:2608.23507v1 Announce Type: new Abstract: Historical people may appear under different languages, scripts, and transcription traditions, while distinct individuals may share highly similar or even identical names. This makes historical identity reconciliation more than a problem of string matching or transliteration. We introduce MHER, a provenance-controlled benchmark for pairwise reconciliation of person-name attestations from the Mongol world. MHER contains a balanced 396-pair Name-only core over 84 primary historical persons and a stricter 160-pair Source-grounded subset constructed from mention-by-source evidence, with entity-disjoint development and test splits. Across five generative systems, correctly Source-grounded evidence improves paired TEST accuracy by 12.96 to 94.44 percentage points relative to Name-only input. On five identical-surface different-person cases, all models fail under names alone (0/25 model-item decisions), whereas Source-grounded evidence yields 24
arXiv:2608.23476v1 Announce Type: new Abstract: Narrow fine-tuning on small, domain-specific datasets can produce broad and surprising changes in model behavior-a phenomenon called weird generalization (WG). Yet, it remains unclear what features of the fine-tuning data are necessary for WG to arise. Here, we address this question by investigating a range of plausibly relevant features, including dataset size, composition, language, presentation style, and novelty relative to a model's parametric knowledge. Further, since WG evaluations rely on small question sets that assess the extent of the generalization, we also analyze how sensitive this measurement is to the set of questions used. Experiments with three open-weight models on four datasets show that the degree of WG (1) depends heavily on dataset composition and language (more than on size); (2) is greater for data familiar from pretraining than for novel data; and (3) is sensitive to the set of evaluation questions used. Collecti
arXiv:2608.23474v1 Announce Type: new Abstract: Vision-language models (VLMs) achieve strong performance on video and image-sequence benchmarks, yet it remains unclear whether they capture temporal structure. To study this question, we formulate temporal grounding as an anomaly detection problem, providing a simple and controlled evaluation that directly tests sensitivity to temporal consistency. We introduce TimeCatch, where temporal anomalies are created by swapping consecutive frames and frame-level anomalies by replacing a frame with Gaussian noise. Models are evaluated on anomaly detection and localization tasks across four synthetic and real-world datasets, alongside a human study. Our evaluation reveals a substantial gap between frame-level and temporal anomaly detection. While VLMs consistently detect frame-level anomalies and often localize them accurately, they perform near chance on temporal anomaly detection and only modestly above chance on localization. Humans, in contras
arXiv:2608.23448v1 Announce Type: new Abstract: This paper presents new Cantonese ParGram resources and evaluates LLMs for knowledge-driven grammar engineering within a controlled experimental paradigm. Using Cantonese ParGram resources as gold standards, with corresponding English baselines, we investigate whether OpenAI's gpt-oss-120b and GPT-5.4 can generate machine-processable grammars from sentences and target formal structures under systematically varied prompting conditions. GPT-5.4 outperformed gpt-oss-120b, while grammars generated from target formal structures generally outperformed those generated from sentences. Although both models could generate locally plausible phrase-structure rules, lexical entries, and templates, they often struggled to coordinate interacting formal constraints, especially in multi-construction settings. The results characterize both the capabilities and limitations of current LLMs for potential integration into AI-assisted expert workflows: LLMs may
arXiv:2608.23421v1 Announce Type: new Abstract: Natural Language Processing (NLP) has grown rapidly over the past decade, driven by digital transformation in the Arab world, social media, and large language models (LLMs). Despite this growth, a comprehensive quantitative meta-analysis of the field remains absent. This study presents a large-scale bibliometric and topic-based analysis of 7,120 Arabic NLP papers published between 1960 and 2026, sourced from six collections. We employ BERTopic for topic modeling, regression analysis to identify citation predictors, social network analysis for co-authorship structures, and geographic mapping. Our findings show a significant publication surge after 2020, driven by transformer models and LLMs. Topic modeling identifies 19 substantive themes, the largest centered on text, speech, translation, and recognition. Citation analysis reveals a positive correlation between paper age and citations (r = 0.245, p < 0.001); regression shows that indexing
arXiv:2608.23411v1 Announce Type: new Abstract: LLM value studies often merge questionnaire ratings, pairwise choices, and values inferred from generated text into one profile. That merge assumes that the three observations describe the same stable preference. STONIC tests this assumption on 5,144 situations from four banks and 35 fixed model configurations. It compares responses rated in isolation, choices made under counterbalanced conflict, spontaneous answers, and later choices between a model's own answer and authored alternatives. 10 of 17 configurations with usable behavioral data preserve the endorsement-choice relation across banks. Every one of the 17 eligible configurations prefers its own earlier answer (median effect 0.790), although option position changes the choice rate in every eligible configuration. Profile shape transfers most strongly from ratings to conflict choices and weakens for spontaneous text. Three-way annotation of 200 L3 responses provides a task-local ch
arXiv:2608.23391v1 Announce Type: new Abstract: Structured data exists in many forms (tables, knowledge graphs, charts, and time series), and converting it into text may involve different generation tasks. However, most prior work on data-to-text (D2T) generation has focused on specific tasks and datasets, relying either on task-specific training data or on the zero-shot capabilities of large language models. We study cross-domain D2T generation in a setting where neither in-domain training text nor test references are available, and where domains, generation goals, and input structures vary substantially. We compare data-driven knowledge distillation (DDKD) against zero-shot inference and fine-tuning on out-of-domain D2T data, and introduce structure-preserving augmentation via structural subsampling and perturbation. Experiments on five benchmarks show that, at constant model size (1.7B parameters), DDKD consistently outperforms both fine-tuning and zero-shot inference. Moreover, the
arXiv:2608.23390v1 Announce Type: new Abstract: English Wikipedia is often treated as the default encyclopedic source, yet non-English Wikipedia editions can contain richer locally grounded information for long-tail figures. We study cross-lingual biography enrichment: enriching an existing English biography with facts supported by a non-English biography about the same person. Focusing on women from non-English-speaking contexts, we introduce \textsc{CLAW-4L}, a benchmark consisting of 300 Wikipedia biography pairs linking an English biography with its French, Chinese or Azerbaijani counterpart, along with claim annotations and a fine-grained claim-pair relation corpus. We propose a claim-based enrichment framework that extracts English claims from both biographies, aligns them to identify enrichment evidence from the non-English biography, and rewrites the English biography using the selected claims. Our results show that non-English Wikipedia biographies provide valuable evidence fo
arXiv:2608.23358v1 Announce Type: new Abstract: The performance gap between low- and high-resource languages in LLMs is widely known, but it remains unclear which internal model factors drive these disparities. In this paper, we characterise this gap through the lens of representational geometry. Comparing the geometric properties of hidden representations across 30 languages reveals that LLM geometry is systematically related to language data availability. The most consistent effect is in final layers, where low-resource languages exhibit representational degeneration. To counter this, we investigate the effectiveness of regularisation terms to penalise degeneration during continued pretraining (CPT). Experiments monolingually adapting 9 base LLMs to 10 African languages show that geometric regularisation successfully reduces representational degeneration during CPT. For larger models, cosine similarity-based regularisation marginally improves performance over vanilla CPT, with more c
arXiv:2608.23353v1 Announce Type: new Abstract: Mixed-integer programming (MIP) lies at the core of operations research and industrial optimization. While large language models (LLMs) have recently shown promise in automated MIP modeling from natural language, they prioritize semantic correctness but overlook formulation strength, severely bottlenecking the efficiency of downstream solvers. We propose FormuEvo, an LLM-guided evolutionary framework for automated discovery of solver-efficient MIP formulations. FormuEvo frames MIP formulation design as evolutionary optimization over the symbolic space of MIP formulations, represented as executable modeling programs, by iteratively generating, evaluating, and selecting stronger candidates via LLM-driven crossover, mutation, and repair operations. To move beyond blind exploration, FormuEvo introduces a solver-informed diagnosis mechanism that exploits fine-grained solver statistics as verbal gradients for targeted refinement. Additionally,
arXiv:2608.23338v1 Announce Type: new Abstract: LoRA fine-tuning is standard for adapting LLMs to reranking, but it remains unclear where in the network task-specific relevance behavior is learned and what attention-level changes accompany that learning. Through ablation and attention experiments, we identify where LoRA attention updates to RankLLaMA improve performance and whether those gains coincide with interpretable relevance-oriented attention patterns such as lexical matching, rarity sensitivity, and query-document interaction. We find that given LoRA fine-tuned MLPs throughout the network, restricting LoRA attention updates to a compact mid-network region is sufficient for recovering over half of the performance gained by applying LoRA to all attention layers, and that omitting attention fine-tuning in this region hurts performance more than elsewhere in the network. Additionally, we show that regions where applying LoRA affects performance the most overlap with regions where f
arXiv:2608.23327v1 Announce Type: new Abstract: Flesch Reading Ease (FRE) and the Flesch-Kincaid Grade Level (FKGL) are widely used readability scores for English computed from the same two document statistics, yet their stability on long documents need not imply invariance to lexical composition. Surprisingly, under a topic model with an explicit sentence-boundary token, both scores converge almost surely to deterministic functions of the document topic distribution through just two scalar rates: in the long-text limit, all score variation is mediated by topical composition rather than any residual readability signal. The theory covers both formulae, while the experiments evaluate FKGL. In a fixed admixture with rank[1, q, s] = 3, fibres through interior topic vectors are locally (K-3)-dimensional, whereas regular iso-score level sets are locally (K-2)-dimensional and curved. In out-of-fold evaluation on two balanced corpora, Brown and the written BNC, a topic vector inferred from one
arXiv:2608.23311v1 Announce Type: new Abstract: Policy optimization (PO) for Large Language Models faces a stability--exploration trade-off, currently mediated by an action-side Policy-KL regularizer. This puts practitioners in a double bind: keeping Policy-KL constrains response behavior and consumes the action-side exploration budget, while dropping it leaves the optimization without an explicit drift control. We argue for an alternative that breaks the dilemma by moving regularization to the input side. As training progresses, the distribution over training queries induced by the current policy drifts unchecked from its pre-RL reference distribution. Concretely, Environment-Regularized Policy Optimization (ERPO) introduces a Query-KL (QKL) term that bounds this query distribution shift, together with a dataset-static reference-derived per-query weight that biases each per-query update toward queries typical under the reference. The QKL gradient flows strictly through the query likel
arXiv:2608.23284v1 Announce Type: new Abstract: Dynamic Embedded Topic Models (D-ETM) provide an interpretable framework for modeling temporal semantic evolution, but cross-corpus comparison remains difficult because topics are often learned independently and aligned only after training, a process that does not guarantee stable topic correspondence across corpora and time. To address this problem, we propose a D-ETM framework that first learns a common dynamic topic space over a merged multi-corpus collection, which we call the shared backbone, then introduces corpus-specific residual adaptation around the frozen backbone without creating separate latent topic spaces. This design preserves a shared topic index for cross-corpus comparison while allowing each corpus to specialize lexically. We evaluate the framework on three temporally structured corpora spanning 97 years: the Corpus of Historical American English, Harvard Business Review, and International Labour Review. Residual adapta
arXiv:2608.23265v1 Announce Type: new Abstract: In long-term collaboration spanning multiple meetings, factual states such as decisions and risks are continually revised, overturned, and replaced. Existing long-context methods typically stack the entire history, while many RAG and structured-memory methods organize knowledge as static or append-only facts and rely on semantic relevance at read time. Without explicit modeling of knowledge lifecycles, these approaches may retain conflicting old and new states simultaneously or discard history, leading to stale retrieval and answers that are difficult to verify. We present EvoWiki, an incremental question-answering architecture for dynamic long-form text. EvoWiki decouples offline incremental construction (BUILD) from online structured reading (READ). BUILD captures the intra-meeting micro-evolution from proposal to decision and uses entity version chains and a fine-grained State-Overwrite Protocol to explicitly distinguish current valid
arXiv:2608.23261v1 Announce Type: new Abstract: Event extraction is fundamental to information extraction. Prior approaches often separate event detection and argument extraction or depend on dataset-specific designs, limiting scalability and cross-domain generalization. We propose a unified generative sequence-to-sequence framework that performs event extraction subtasks jointly and supports both pipeline and end-to-end configurations. We fine-tune pretrained language models on multiple event datasets across diverse domains, enabling a single model to retain domain-specific semantics while generalizing over large and evolving label spaces. We demonstrate these capabilities through a web-based application tailored for researchers and practitioners. The platform supports document upload, schema-aware event extraction, visualization of triggers and arguments, and comparison of different extraction configurations across domains.
arXiv:2608.23248v1 Announce Type: new Abstract: Traditional clinical prediction models rely on task-specific pipelines and curated, structured data, which scale poorly and underutilize unstructured text. To address this, we introduce future querying, a paradigm that probes whether large language models (LLMs) can function as implicit medical world models by evaluating their ability to answer time-indexed clinical queries about a patient's future. Our framework operates on unstructured clinical documentation using endpoint-agnostic training, enabling a single model to answer diverse clinical queries over patient trajectories without manual feature engineering or task-specific retraining. We show that small, locally fine-tuned open-weight models can match or approach larger proprietary systems, making the framework suitable for privacy-preserving, on-premise deployment. Evaluated on a new synthetic medical reports dataset and real ICU notes from the MIMIC-IV dataset, our results provide
arXiv:2608.23244v1 Announce Type: new Abstract: Large language models (LLMs) often produce fluent but incorrect answers with unwarranted confidence. A central limitation is that standard LLMs represent uncertainty through a single predictive distribution, conflating epistemic ignorance with genuine ambiguity. We introduce Credal Large Language Models (CLLMs): an ensemble of LoRA adapters induces a credal set whose lower and upper probabilities expose the spread of plausible predictive distributions rather than collapsing to a single softmax output. From this representation we derive two complementary commitment scores. Credal Token Commitment (CTC) is a token-space score that combines lower-bound support, credal width, and intersection entropy, computed without additional generation. Semantic Commitment Consistency (SCC) extends commitment to semantic space using sampled completions, with SCC-Gap measuring the mismatch between token-level and semantic-level support. We evaluate halluci
arXiv:2608.23235v1 Announce Type: new Abstract: Event extraction aims to identify event triggers, classify event types, and extract arguments to construct structured event representations. Despite strong in-domain performance, developing models that generalize robustly across domains remains challenging due to variations in contextual expressions and event schemas. Prior unified and multi-task approaches improve in-domain accuracy but exhibit limited flexibility when applied to unseen domains. Even large language model-based methods that provide full event ontologies at inference time often underperform compared to smaller, task-specific fine-tuned models. We propose a unified multi-domain and multi-task training framework that models heterogeneous event schemas within a single model. Our approach introduces domain conditioning signals, jointly with task-specific prompts, enabling dynamic adaptation to dataset-specific schemas without requiring complete event label sets at inference ti
arXiv:2608.23214v1 Announce Type: new Abstract: Biomedical knowledge exists in two complementary but distinct forms: unstructured scientific literature and structured knowledge graphs (KGs). Aligning them is essential for knowledge grounding, evidence retrieval, and KG completion, yet existing methods do not explicitly align free-text evidence with KG triples. We present a unified framework for systematically studying design choices for aligning biomedical text and KGs. With a text encoder and a KG embedding model both frozen, we learn only a lightweight projection between their spaces via a contrastive objective. This enables a fair comparison across six design dimensions: text encoder, KG embedding model, projection head, triple composition, training direction, and hard-negatives sampling. We construct CTD-Align, a corpus of over 22K one-to-one tripledocument pairs linking chemical-gene interactions from the Comparative Toxicogenomics Database to supporting PubMed passages. We evalua
arXiv:2608.23200v1 Announce Type: new Abstract: Large language models are increasingly expected to execute complex workflows whose success depends on maintaining interdependent constraints and producing artifacts that satisfy strict end-to-end verification. Yet successful execution experience is typically lost after a single run, forcing subsequent models to rediscover strategies and failure modes from scratch. We study whether such experience can instead be externalized and reused through EvoMap, where verifier-confirmed execution trajectories are consolidated into structured Gene. To evaluate this setting, we introduce the Long-Workflow Benchmark (LongWoF-Bench), comprising 778 machine-verifiable tasks across code generation, agent-environment synthesis, mathematical reasoning, and rule following. On the 252 tasks with verifier-confirmed Opus trajectories, evolved EvoMap Gene outperform Skill across all seven evaluated models by 8.7-15.5 percentage points, with the gains extending to
arXiv:2608.23172v1 Announce Type: new Abstract: Large-scale vision-language models (VLMs) have demonstrated remarkable versatility across a wide range of multimodal tasks. However, understanding humor remains challenging because humorous content often depends on subtle interactions among entities, events, context, and implicit relationships across image and text modalities. These interactions can involve complex chains of reasoning that are difficult to capture through conventional prompting or linear chain-of-thought reasoning. In this work, we propose CaRGo-T (Causal Reasoning Graph-of-Thought), a reasoning framework that represents the causal and contextual relationships underlying multimodal humor as a lightweight graph-based reasoning structure. The graph is serialized into a code-based representation generated by a VLM, which can subsequently be interpreted by the same or a different VLM to produce the final prediction in zero-shot or in-context learning settings. We evaluate CaR
arXiv:2608.23167v1 Announce Type: new Abstract: Diffusion Language Models (DLMs) exhibit strong parallel decoding capabilities by denoising multiple tokens in a single generation step. However, this parallelism comes with substantial computational overhead, as each step requires interactions with all suffix tokens. Existing methods typically reduce this cost by retaining only a local suffix window as a substitute for the full suffix. Despite their effectiveness, these methods overlook the structural heterogeneity across suffix regions and re-initialize suffix tokens with identical representations at each timestep. To this end, we propose a structured suffix modeling method for efficient DLM inference. Specifically, we divide the suffix into three regions, i.e., the local, middle, and tail regions, and retain different numbers of suffix tokens in each region according to their structural roles. Moreover, we incorporate the decoding results from the previous step into the suffix token re
arXiv:2608.23152v1 Announce Type: new Abstract: Counterspeech effectively neutralizes the impact of online hate. Although prior work explores automated counterspeech generation, it largely emphasizes stylistic control while treating hate speech as homogeneous, overlooking that distinct forms of abuse require fundamentally different counterspeech strategies. To address this gap, we introduce FIRE (Factuality Informed Multi-Agent Reasoning Framework) that first decomposes hate speech into one of the five distinct categories (misinformation, stereotype, conspiracy, dehumanizing, non-factual), and then maps it to a targeted counterspeech style. To facilitate FIRE, we curate FactualCS, a novel dataset of $4,784$ instances that provides the annotations regarding hate categories, reasoning traces, and evidence mappings, which are critical elements for grounded generation that are missing in prior work. A comprehensive evaluation across $28$ baseline configurations demonstrates that FIRE signi
arXiv:2608.23149v1 Announce Type: new Abstract: The alignment of Large Language Models heavily relies on English-centric high-quality preference data, which often leads to suboptimal performance in other languages. In this paper, we propose Cross-Lingual Ranking Preference Optimization (CRPO), a novel framework that leverages robust preference knowledge from English to facilitate preference alignment in the target language. We design a hierarchical structure within parallel preference pairs across the target language and English to jointly optimize intra- and inter-lingual preferences, thereby enhancing language adaptation and output quality. Building on the LambdaLoss framework, CRPO goes beyond the binary comparison based optimization by providing a relative ranking signal across multiple candidate responses. Our experiments across five languages with varying resource scales demonstrate that CRPO consistently outperforms standard approaches in both instruction-following and knowledge
arXiv:2608.23124v1 Announce Type: new Abstract: 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
arXiv:2608.23120v1 Announce Type: new Abstract: Pnar, an Austroasiatic language spoken by approximately 0.4 million people in the Jaintia Hills of Meghalaya, lacks the digital corpora and natural language processing (NLP) resources. This paper presents the first machine translation study for the English and Pnar language pair. Using articles collected from the Wyrta newspaper, we built a parallel corpus comprising of 10,234 sentences and trained phrase-based statistical machine translation (SMT) systems the models using 9,563 parallel corpora under three configurations for each direction using Moses, GIZA++ , KenLM, varying lexicalized reordering and minimum error rate training (MERT) tuning. The models are evaluated on a held out test set of 371 sentences, the best performing system achieves a BLEU score of 14.97 (chrF2: 33.42, TER: 77.60) for Pnar to English and 11.16 (chrF2: 31.38, TER: 93.51) for English to Pnar, establishing the first quantitative benchmark for this language pair.
arXiv:2608.23104v1 Announce Type: new Abstract: Molecular science represents an important frontier for LLM-based agents. Unlike general agents that mainly operate over natural language, code, or web environments, molecular LLM agents must perceive, reason about, and act upon chemical objects across symbolic strings, molecular graphs, 3D conformations, spectra, simulations, and wet-lab measurements. Their capabilities depend on chemically faithful molecular perception, an LLM-centered agent framework, domain-specific tool grounding, and computational or experimental feedback, in addition to planning and tool use. This work develops a conceptual framework for molecular LLM agents from two complementary perspectives. First, we introduce an architectural view of molecular-agent design, covering molecular representation and perception, the agent framework, domain-specific toolboxes, and learning and optimization. Second, we propose a scientific autonomy ladder inspired by staged autonomy in
arXiv:2608.23095v1 Announce Type: new Abstract: Media bias detection relies on definitions and examples that specify what counts as bias, yet these specifications often vary across datasets or remain implicit, even when given the same name. Such variation makes it unclear whether models trained for the same bias category learn the same construct or different phenomena, a problem largely overlooked in prior work. We examine how definition choice affects bias annotation in a between-subjects experiment with 354 participants and a parallel evaluation with four LLMs. Participants and models rate six news articles across four bias categories using definitions that vary in conceptual framing and elaboration. Across 8,496 human and 28,800 LLM ratings, we find that the conceptual target of a definition drives annotation divergence, while construct-preserving elaboration does not: conceptual framing significantly shifts annotations for humans and does so even more strongly for LLMs. We discuss
arXiv:2608.23067v1 Announce Type: new Abstract: Agent Skills are reusable procedural modules that are increasingly injected into coding-agent sessions to encode framework conventions, anti-patterns, and reusable tools. However, because each injected Skill expands the prompt of every query, an effective Skill benchmark must determine not only whether an agent can solve a task, but whether the Skill should have been injected at all. We introduce WebDev-Skills-Bench and use it for a controlled empirical study of 31 public WebDev Skills on 50 Web-Bench projects and 1,000 ordered tasks. The benchmark compares four matched conditions, including a length-matched irrelevant control and leave-one-out component ablations. To isolate Skill effects from prompt-length artifacts, we place only SKILL.md in the prompt while mounting auxiliary files into the agent workspace. Across four models, target Skill injection reduces mean Pass@2 by 1.3% to 4.2%, lowers task completion depth, and increases token
arXiv:2608.23047v1 Announce Type: new Abstract: Misinformation that cites legitimate papers can be especially harmful when it distorts what those studies actually report. While existing automatic fact-checking systems based on large language models (LLMs) can assess whether a model assigns an Incorrect verdict and can gen- erate explanations for that decision, they typi- cally do not indicate whether the model follows the same reasoning path as human experts or arrives at the verdict through a different but still valid path. In this work, we introduce a graph- based framework (typed reasoning graph) for comparing human and LLM reasoning paths in scientific fact-checking. Building on prior work on fallacious reasoning in biomedical misinformation, MISSCIPLUS (Glockner et al., 2025), we model each explanation as a rea- soning graph that links the false claim to the relevant study context, study findings, fallacy- supporting premises, and fallacy labels. This representation enables one-to
arXiv:2608.23037v1 Announce Type: new Abstract: This paper introduces the Multilingual FrameNet Corpus (mFNC), a novel resource that extends the English Berkeley FrameNet corpus by collecting and harmonizing existing language-specific corpora across nine additional languages: Brazilian Portuguese, Chinese, Dutch, French, German, Italian, Korean, Latvian and Swedish. By training models that rely on different architectures on the mFNC, we consistently outperform existing state-of-the-art Frame Semantic Parsers in both multilingual and cross-lingual settings, underscoring the importance of multilingual training data. The mFNC and our trained FSP models are openly available at https://github.com/beatrice-f/mFNC.
arXiv:2608.23029v1 Announce Type: new Abstract: Multi-agent debate can improve large language model reasoning by eliciting diverse hypotheses and critiques, yet its performance is often constrained by weak moderation. Common pipelines rely on fixed budgets, agreement-based stopping, or untrained judges, leading to redundant deliberation and unreliable evidence aggregation. We cast moderation as a meta-cognitive process, monitoring debate utility, controlling deliberation, and adjudicating a final answer, and introduce Meta-Moderator, a learnable framework that dynamically regulates debate and decides when to finalize an answer. Meta-Moderator is trained independently of the debaters via outcome-driven policy optimization, making debate regulation an explicit capability rather than an incidental effect of prompting. Across five benchmarks, Meta-Moderator outperforms widely used decision layers and transfers across tasks and system configurations. Further analyses show that it allocates
arXiv:2608.23026v1 Announce Type: new Abstract: Multilingual LLM outputs can vary across sociocultural contexts. However, evidence of cultural grounding can be misleading: identity labels may be inferred from explicit or indirect textual cues, while names and wording can reveal the source language. Treating all these signals as evidence of cultural grounding may obscure potential biases. We present a human-validated, multi-agent audit that separates three questions: whether outputs reproduce social biases, whether identity groups are represented differently, and whether outputs reflect cross-cultural patterns. The study analyzes 89,253 outputs from 12 LLMs in English, French, and Chinese, spanning 18 occupations and three task conditions. We find that bias representation varies systematically across languages and tasks. Removing direct identity cues sharply reduces identity-label prediction in English and Chinese, but has a much smaller effect in French. Across all language-genre setti
arXiv:2608.23023v1 Announce Type: new Abstract: An LLM router picks which model should answer each query. The appeal is that models fail on different questions. Whatever single model is best overall still gets some wrong, and another model in the pool gets many of those right. Getting that choice right every time is the ceiling, and a router is an attempt to approach it. However, recent work reports that routers do not get close. Across 21 routing methods on five benchmarks, sharply different designs land within a fraction of a point of each other, and all of them stay far below that ceiling. Learned routers often fail to beat simply always calling the strongest model. We ask what those missed questions have in common. We set fourteen models to answer all 294 questions, with 7 task types across 3 languages: Korean, English and Hindi. We ran the whole matrix twice, changing nothing, but 5.37% of the 4,116 model-question pairs came out scored differently anyway. Run-to-run movement like
arXiv:2608.23020v1 Announce Type: new Abstract: Large language models (LLMs) require effective unlearning to address privacy regulations and safety concerns. However, achieving precise forgetting without compromising general utility remains challenging. Existing sequence- and token-level methods penalize target outputs without modeling their context-dependent retrieval paths, which can disrupt linguistic structure or suppress benign knowledge. We present ADU, a fine-grained, training-based framework that shifts unlearning from token erasure to contextual attention-pathway decoupling. Exploiting the functional distinction between local and global attention heads, ADU identifies preplan positions that retrieve persistent sensitive anchors and fixes their candidate paths under the original model. It then trains attention-projection adapters to suppress attention mass along these paths while preserving local-attention structure and retain-set language modeling. Post-training activation exc
arXiv:2608.22985v1 Announce Type: new Abstract: Activation steering is often evaluated under the answer encoding used to construct the direction. A reported gain may reflect the intended judgment or compatibility with answer identifiers seen during construction. We introduce Cross-Encoding Steering Evaluation, which freezes an intervention while re-encoding answers to the same held-out items. On NormBank, after A/B/C identifiers are reassigned, contrastive activation addition (CAA) induces larger target-versus-source score changes for the extraction indices than for the semantic labels under the new mapping. We call this extraction-index following. Varying identifier vocabulary (A/B/C, X/Y/Z, or 1/2/3) and row order shows that the effect tracks extraction index rather than row position. After matching direction norms across layers, extraction-index following emerges mainly at later depths. A low-rank output-sensitive component containing 15.4% of the direction's squared norm retains 96
arXiv:2608.22967v1 Announce Type: new Abstract: Practical molecular inverse design is rarely a one-shot generation problem; it often takes the form of closed-loop candidate-pool enrichment, where under a limited oracle budget the goal is to \emph{increase the fraction of generated molecules that match a desired property profile}. Bayesian optimization (BO) offers a natural framework for this setting, yet standard Gaussian-process surrogates typically operate in compressed continuous embeddings, which discard the substructural and reference-similarity signals that chemists naturally use to decide where to look next. We propose \textbf{\method}, a closed-loop framework in which the surrogate, rather than the generator, is treated as the locus of design choice, and instantiate it with a frozen large language model that reasons directly over the task instruction, SMILES-level optimization history, and oracle feedback in their native textual form. At each iteration, the surrogate returns a
arXiv:2608.22956v1 Announce Type: new Abstract: Concept-bottleneck controllable generation routes multi-attribute control through a low-dimensional concept code that, at deployment, must be synthesised from a target attribute configuration. We study this problem in concept-bottleneck text generation under multi-axis compositional generalisation, comparing three ways to obtain the inference-time code: classifier inversion against the encoder heads, reference-text encoding, and a post-hoc label-conditioned prior. Since a concept code admits no direct LM-fluency term, regularising inversion must instead constrain the code toward the encoder's training distribution. We therefore test bare inversion and three regularised variants: label-agnostic and label-conditioned Mahalanobis penalties, and a conditional normalising-flow density baseline. Every inversion variant we test underperforms a simple post-hoc prior fitted to per-combination encoder means on the same checkpoints, across three bac
arXiv:2608.22948v1 Announce Type: new Abstract: Large language models are increasingly used to propose research ideas, yet the prevailing ways of judging such ideas supply no shared decision rule: free-form judging sways with style and position, and scoring against a later paper rewards recovery of one realized trajectory. We introduce a benchmark that carries a proposal from Literature to Test: the Lit2Test benchmark centers on a six-field contract organized around a falsifying outcome, so that every proposal precommits the observation that would prove it wrong, making its quality decidable in the first place rather than merely arguable. Built prospectively from 200 real-paper neighborhoods, Lit2Test elicits proposals from four frontier models and compares them through 1,200 pairwise comparisons judged blind in both presentation orders. The protocol audits its own reliability through diagnostic controls and bounded human calibration, with three annotators corroborating the conclusions
arXiv:2608.22922v1 Announce Type: new Abstract: We present HelaBERT, a family of two BERT-based masked language models pre-trained from scratch on approximately 1 billion tokens of Sinhala text sourced from MADLAD-400, CulturaX, and a custom corpus comprising news articles, Sinhala Wikipedia, and web crawl data. HelaBERT-Small (~23.3M parameters, 6 layers) and HelaBERT-Large (~110M parameters, 12 layers) both use a SentencePiece Unigram tokenizer (vocabulary size 32,000) tailored to Sinhala's agglutinative morphology and complex script. We evaluate both models on four downstream Sinhala text classification tasks: news category classification, news source classification, sentiment analysis, and writing style classification, using 5 independent seed runs with stratified 80/20 train/test splits. We additionally propose a dual pooling classification head and evaluate it systematically across all four tasks, finding consistent improvements on sentiment analysis and a moderate gain on news c