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:2607.15178v1 Announce Type: new Abstract: Transformer reasoning is limited by autoregressive decoding, which repeat edly compresses rich hidden computation through token space and makes it difficult for intermediate reasoning states to persist across time. We in troduce Transformers with Temporal Middle-Layer Recurrence (T2MLR), a transformers-based latent reasoning architecture that fuses a cached middle layer representation from the previous token directly into an earlier layer of the current token position, enabling abstract intermediate computation to persist across decoding steps with little inference overhead. Across natural-language pretraining and multi-hop reasoning finetuning, T2MLR consistently outperforms data- and parameter-matched Transformer base lines. Moreover, applying recurrence to only a localized middle-layer block (as little as 20% of the network) often outperforms full-layer recurrence. Im portantly, T2MLR does not require pretraining from scratch: retrofit
arXiv:2607.15175v1 Announce Type: new Abstract: Whether neural language models (NLMs) possess the ability to distinguish strings on the basis of their grammaticality remains a debated topic in the computational linguistics literature. Existing evidence has largely relied on probability-based measures, testing whether models assign higher probabilities to grammatical than ungrammatical strings. However, probability comparisons have been criticized as a measure for grammatical knowledge based on the assumption that grammaticality is inherently entangled with likelihood. Model-assigned probability is a function of many related sentence properties, such as lexical frequency, plausibility, and world knowledge. In this work, we move beyond probability-based evaluations and investigate whether grammaticality is encoded in the internal representations of NLMs. Using mass-mean probing, we test whether grammatical and ungrammatical sentences are systematically separated in representational space
arXiv:2607.15095v1 Announce Type: new Abstract: The formation of political coalitions is a complex negotiation driven by both concrete policy objectives and deep-seated ideological convictions. While Large Language Models (LLMs) open new avenues for computational political science, the neutrality and helpfulness biases instilled by Reinforcement Learning from Human Feedback (RLHF) prevent them from sustaining steadfast partisan behaviour. We present a multi-agent framework that reconciles factual grounding with ideological alignment by combining Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Retrieval-Augmented Generation (RAG): DPO instils aggressive party-specific personas, while a per-party RAG pipeline keeps each agent bounded to its official manifesto. We operationalize the framework on the 2019 Flemish election, deploying the partisan agents in a hub-and-spoke negotiation arbitrated by a formateur. To make the emergent negotiation interpretable, we introd
arXiv:2607.15092v1 Announce Type: new Abstract: Rubrics provide structured, fine-grained signals for training and evaluating large language models (LLMs). Yet reliable query-specific rubrics are difficult to construct. Existing approaches often derive supervision from human-written rubrics, preference data, or sampled responses. Direct query-to-rubric generation avoids these resources, but provides no explicit check that a plausible rubric is useful. Such a rubric may fail to distinguish answer quality, reward an optional style, or penalize a valid alternative strategy. We introduce Rubrics on Trial, a query-only framework that evolves a rubric set from an empty set without external annotations or model training. It derives supervision solely from synthetic rubric-conditioned response pairs and validates each proposed rubric before adding it, screening out non-discriminative, over-specific, and style-only candidate rubrics. Experiments across five preference benchmark suites demonstrat
arXiv:2607.14989v1 Announce Type: new Abstract: Large language models are increasingly evolving from text generators into general agents capable of understanding user requests, invoking external tools, and completing complex tasks through interaction. However, existing agent benchmarks often focus on limited scenarios, tool ecosystems, or interaction formats, making it difficult to systematically characterize model capabilities across heterogeneous application settings. We introduce OmniaBench, a benchmark for evaluating general agents across diverse scenarios with explicit state spaces. We derive application-oriented scenario knowledge from app stores, product documents, industry resources, Web retrieval, and human refinement, forming a hierarchical taxonomy that spans ToC, ToB and ToE with 90 level-1 and 354 level-2 domains. Based on this taxonomy, we construct executable environments and synthesize single-turn and multi-turn tasks through four complementary routes: DAG, DAG-S, Solve
arXiv:2607.14967v1 Announce Type: new Abstract: Most existing approaches to AI-Generated Text Detection (AIGTD) treat documents as static objects and base their decisions on aggregate statistics or globally compressed embeddings. However, this perspective overlooks the inherently dynamic nature of autoregressive generation, where content evolves progressively through the latent space. In this paper, we reformulate AIGTD as the problem of distinguishing between latent generation trajectories. Instead of relying on static representations, we model how textual representations evolve across the sequence. To this end, we propose Geometric Trajectory and Contrastive Learning (GTCL), a framework that segments the document into ordered local units, encodes each unit in an embedding space, and constructs a structured and sequence-level representation. GTCL then applies contrastive learning to these trajectories to learn geometric regularities associated with the autoregressive generation. Evalu
arXiv:2607.14905v1 Announce Type: new Abstract: Given the current trend to employ large language models (LLMs) in almost any imaginable context, LLM-generated text detection and authorship attribution have become a pressing issue. Prior work has primarily focused on surface-level linguistic features, an approach shown to be susceptible to paraphrasing and other obfuscation techniques. In this paper, we go beyond the linguistic surface, extracting and analysing reasoning structures in LLM-generated texts with the goal of capturing more complex signals of LLM authorship. We propose a graph neural network approach that leverages reasoning graphs extracted by an argument mining pipeline, demonstrating improved robustness and generalisation over a traditional Longformer baseline. Our approach outperforms the baseline by up to 27 percentage points under the obfuscation attacks such as paraphrasing and backtranslation, and 19 percentage points when evaluated on the texts generated by the unse
arXiv:2607.14777v1 Announce Type: new Abstract: Large language models are increasingly trained as interactive agents for long-horizon tasks involving multi-turn interaction, tool use, and environment feedback. Outcome-based reinforcement learning (RL) provides a practical optimization paradigm, but its sparse trajectory-level rewards offer limited guidance on intermediate decisions, leaving a supervision gap between episode-level outcomes and token-level policy learning. We propose SEED (SElf-Evolving On-Policy Distillation), a self-evolving framework that converts completed on-policy trajectories into training-time hindsight skills and distills their behavioral effect back into the policy model. SEED first fine-tunes the policy to analyze completed trajectories and generate natural-language skills that capture reusable workflows, decisive observations, or failure-avoidance rules. During RL, the current policy both collects trajectories and serves as the analyzer that extracts hindsigh
arXiv:2607.14769v1 Announce Type: new Abstract: Existing text summarization research has focused much on monologic information (e.g., newspaper articles, reports) without accounting for the interaction between speakers or authors. In contrast, dialogues are a rich communication channel where multiple participants conduct back and forth exchanges to construct meaning. We propose a dialogue summarization framework that explicitly models both semantic and emotion dynamics using multimodal dialogue inputs, built on an adapted hierarchical Chain-of-Agents approach. We decompose dialogues from two perspectives: (1) topic segments based on the utterances of all participants, and (2) participant-specific utterance segments. These are used to generate corresponding summaries while incorporating automatically inferred emotions. Topic- and participant-level summaries are aggregated into a dialogue summary capturing semantic content and emotion trajectories. To evaluate beyond content accuracy, we
arXiv:2607.14735v1 Announce Type: new Abstract: Transparent educational question answering asks for answers that are not only correct but explainable, and doing so with small models rules out the reasoning power of the largest proprietary systems. The EXACT 2026 competition poses this problem concretely: open-weight language models of at most 8B parameters, self-hosted, with a natural-language explanation for every answer. It pairs two tasks: logical reasoning over university regulations, and multi-step physics problem solving. We describe the system that team \cotu{} developed to address both, a neuro-symbolic Program-of-Thought pipeline in which a 4B backbone writes a program rather than stating an answer directly: for regulation queries it emits a Z3 encoding whose entailment verdict grounds the deduction, and for physics it emits numerical Python, both wrapped in a shared self-correction loop and a unified explained-JSON output. Answer-type routing, distillation-based task fine-tun
arXiv:2607.14709v1 Announce Type: new Abstract: Financial question answering over hybrid tabular and textual data may require multi-source reasoning and precise numerical computation. While large language models (LLMs) can generate intermediate reasoning steps, natural-language rationales remain prone to arithmetic errors, making them an unreliable supervision source for distillation. Building on programmatic distillation, we develop an approach that transfers reliable numerical reasoning from a large teacher model to a compact student using execution-verified Python programs instead of free-form textual rationales. It leverages gold derivations to guide teacher-side program synthesis and retains only programs that execute correctly and produce the gold answer, ensuring high-quality supervision. We further introduce an iterative recovery stage that revisits teacher-failed examples, enabling the student to recover and incorporate newly verified programs into training. Experiments on TAT
arXiv:2607.14707v1 Announce Type: new Abstract: Large language models routinely produce fluent answers to single-shot prompts, yet deploying them as reliable components of a domain decision system is substantially harder. Closing this gap is the work of harness engineering: the deliberate composition of deterministic scaffolding (symbolic filters, retrieval, schema-typed I/O, LLM-as-judge loops, HITL gates, persistent state, audit trails) around an LLM core. We present a case study in academic supervision, a domain combining high-stakes recommendation, longitudinal accountability, and structured operational workflows. We compare a baseline (ASA), a GPT-5 chatbot with no scaffolding, against a multi-module system (ASuS) that wraps the much smaller GPT-4o-mini in a LangGraph harness with symbolic-semantic retrieval, schema-validated outputs, LLM-as-judge with bounded retry, HITL gates, deterministic weighted risk scoring with LLM narration, and a per-node SQLite audit trail. The evaluati
arXiv:2607.14647v1 Announce Type: new Abstract: Speculative decoding accelerates large language model (LLM) inference without compromising output quality. Recent parallel drafting methods further improve single-request performance by decoupling draft length from drafting latency, enabling longer drafts and higher mean accepted tokens (MAT). However, under high request concurrency, long drafts waste substantial computation on rejected tokens, increasing verification cost and potentially making speculative decoding slower than autoregressive decoding. We present D-Cut, an adaptive pruning method that selects draft tokens jointly across the batch and concentrates the verification budget on tokens most likely to be accepted. D-Cut is motivated by two observations. First, acceptance lengths vary considerably across concurrent requests; D-Cut therefore performs cross-request pruning, allocating the verification budget adaptively according to draft confidence. Second, verification cost depend
arXiv:2607.14628v1 Announce Type: new Abstract: Large language models (LLMs) exhibit a well-documented gap between latent capability and consistent activation: the router hypothesis posits that models possess the knowledge to solve a task but lack reliable internal routing to activate it. Prior work in formal mathematical reasoning (SAIR, C\'azares 2026) reports that structural priors (cheatsheets) raise in-distribution performance dramatically, yet collapse below the zero-shot baseline out-of-distribution (OOD) -- and that iterative recalibration amplifies rather than corrects the collapse. We test whether this phenomenon is cross-domain by reproducing the SAIR design in source-code security vulnerability detection, evaluating three LLMs (GPT-OSS-120B, Llama-3.3-70B, Gemma-4-31B) across three vulnerability categories (CWE-798, CWE-284, and the non-CWE N+1 anti-pattern) spanning syntactic, contextual, and semantic complexity, then transferring cheatsheet-augmented prompts to real-world
arXiv:2607.14591v1 Announce Type: new Abstract: 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 helpful
arXiv:2607.14561v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated strong reasoning performance, but their tendency to hallucinate limits their reliability in knowledge-intensive tasks requiring up-to-date and grounded information. Combining knowledge graphs (KGs) with LLMs facilitates the use of explicit symbolic knowledge that can be continuously updated without costly fine-tuning, while benefiting from rapidly advancing LLM reasoning. We propose MARS, a scalable knowledge graph question answering (KGQA) approach that requires no model fine-tuning. Rather than relying on open-ended agentic exploration, MARS performs a structured retrieval procedure that links question entities to the KG and iteratively retrieves relevant next-hop information. At each step, MARS decides whether to continue graph traversal or to generate the final SPARQL query, allowing the model to adapt the retrieval depth to the question while keeping the overall pipeline more predictable
arXiv:2607.14552v1 Announce Type: new Abstract: A standard recipe for distilling the reasoning ability of large language models (LLMs) is to sample chains of thought from the model, keep those that reach the correct final answer, and fine-tune on the survivors. When sampling fails, a common fix shows the generator the gold answer and asks it to write a chain that reaches that answer. We show that this second step degrades the training data in a way that correctness filtering cannot catch. We run a controlled experiment that fixes the generator, the problem set, and the correctness filter, and varies only whether the chain is generated under answer-conditioning, the gold answer shown with a request to reach it. Training a strong instruction-tuned reasoning model on its own answer-conditioned chains sharply lowers its verifiable-reasoning accuracy. The loss grows with difficulty, reaching as much as about 27 points on the hardest competition problems. The mechanism is legible in the chai
arXiv:2607.14542v1 Announce Type: new Abstract: Semantic 3D city models provide rich geometric and semantic information, but remain challenging for non-experts and interdisciplinary researchers to access and query due to their complex structures and specialized data formats. To address this issue, we present CityLLM, a framework for natural-language querying of semantic 3D city models alongside complementary urban datasets. The framework combines spatial and graph databases within an LLM-based workflow that supports iterative query refinement and cross-database chaining. We evaluate CityLLM on a CityJSON dataset of Rotterdam (853 LoD2 buildings) using GPT-OSS, Gemini 3.1, and GPT-5.4, along with selected variants, across multiple metrics: answer correctness, visualization correctness, query success, and retry attempts. A total of 54 natural-language queries are curated across four scenarios: spatial, graph, cross-database, and conversational. Results show strong overall performance, wi
arXiv:2607.14528v1 Announce Type: new Abstract: Large language models (LLMs) frequently contradict themselves when the surface form of a logically equivalent question changes. We present a benchmark of 350 question families (1,750 total questions) for Controlled Reformulation Testing (CRTBench) to evaluate logical invariance. In this benchmark, we investigate LLMs' ability to maintain consistent answers across controlled reformulations, which include contrapositive rewriting, double negation, negation flipping, and passive voice. We evaluate several frontier LLMs and observe an accuracy-consistency gap where GPT-5.4-mini achieves $98.9\%$ base accuracy but only $60.3\%$ family-level consistency, while reasoning-optimized o4-mini achieves $96.9\%$ consistency. From our experiments, we observe that failures cluster around logically nontrivial transformations such as contrapositive rewriting ($72.4\%$ for GPT-5.4-mini) and double negation ($84.6\%$), while surface-level rephrasing remains
arXiv:2607.14480v1 Announce Type: new Abstract: LLM evaluators (trained reward models and prompted LLM-as-a-Judge) are routinely validated via pairwise accuracy. In a multilingual setting, this operates under the premise that high pairwise accuracy implies reliable, language-neutral scoring. We show that this assumption does not hold. We conduct experiments with semantically identical instruction-response pairs across 23 languages, and find that multilingual evaluators assign significantly different scores to different evaluation languages. The bias is statistically significant and consistent across eight open-weight evaluators of different architectures and training paradigms, persists in frontier judges, and is strongly correlated with language resource level: lower-resource languages are scored more generously. Meanwhile, these biases are invisible to pairwise accuracy: evaluators achieve above 90% pairwise accuracy, yet have up to 43% difference in acceptance rate across languages
arXiv:2607.14431v1 Announce Type: new Abstract: We report a way to make a frozen small language model both more capable and dramatically cheaper at once, without changing any weights. Verified knowledge is deposited once as a byte-exact key-value (KV) state artifact and later restored, by graft, into a fresh inference context. The restore is bit-exact: under a pinned deterministic configuration, the grafted logits are byte-for-byte identical to a fresh computation (SHA-256 equality), with zero KL divergence and 100% argmax agreement over fifty samples. We show that own-position graft is the unique numerically exact operating point on a model with floating-point rotary encoding, and we verify byte-exactness on two model scales (12B, 31B) and two GPU targets, one through a pre-registered replay. On AIME 2025, a frozen Gemma-4-12B moves from 80.0% to 93.3% once a verified solution library is grafted, above its own 77.5% and its 31B sibling's 89.2% published anchors. On the recurring case,
arXiv:2607.14400v1 Announce Type: new Abstract: This paper describes DS@GT ARC's submission to the CLEF 2026 LongEval Task 4 on Retrieval-Augmented Generation (RAG). In this submission, we examine a divergence between traditional natural language evaluation metrics and citation integrity as applied to RAG QA systems. We evaluate a corrective pipeline using Corrective RAG (CRAG) and CiteFix against baseline and frontier model benchmark RAG QA scores. While frontier models maximized answer relevance and fluency scores, our RAGAs LLM-as-judge diagnostics indicate that frontier models would correctly identify relevant documents without using their context in answer generation. Conversely, by filtering chunks pre-generation and enforcing strict entailment of generated claims to the cited material post-generation, our corrective pipeline marginally improved citation faithfulness and answer grounding. We propose that evaluation of trustworthy RAG QA requires metrics that reward strict answer
arXiv:2607.14385v1 Announce Type: new Abstract: Large language models achieve strong scores on medical benchmarks, yet these benchmarks evaluate each question in isolation, providing no measure of whether a system can distinguish clinically similar presentations requiring different interventions. We introduce MamaBench, the first counterfactual benchmark for maternal and paediatric AI: 434 expert-authored clinical narratives in 217 pairs across 371 pathologies, evaluated via the Bias Trap Rate (BTR), the conditional probability that a model fails the counterfactual given success on the base case. We propose Evidence-Anchored RAG (EA-RAG), a three-stage retrieval method that replaces aggregate similarity with an evidence coverage objective through clinical parameter extraction, coverage auditing, and contrastive sub-queries. Across eight configurations of four frontier LLMs, base accuracy overstates robust accuracy by 16-28 percentage points in every model. EA-RAG achieves 20.3% BTR and
arXiv:2607.14349v1 Announce Type: new Abstract: While Large Language Models (LLMs) excel in many general NLP tasks, their formal reasoning capabilities are often compromised by content effects, demonstrating a measurable bias towards real-world plausibility. In this paper, we present our system for SemEval-2026 Task 11, which evaluates the ability of models to disentangle formal logic from content across 12 languages with and without distractor premises. We address this challenge using mDeBERTa-v3 networks fine-tuned on a synthetic, rule-based dataset of syllogistic schemes to avoid the semantic noise of LLM-augmented data. To explicitly decouple plausibility from logical structure, our training pipeline employs a multi-objective loss function combining Adaptive Group Distributionally Robust Optimization (DRO), a scheduled differentiable bias penalty, and KL-Divergence consistency regularization. Our system achieved #1 ranks and perfect Ranking Scores (100.0) with 0.00% bias and 100.0%
arXiv:2607.14327v1 Announce Type: new Abstract: Efficient long-context inference is not only about reducing memory cost, but also about keeping useful contextual evidence accessible as generation proceeds. However, existing compression-oriented approaches, such as key-value (KV) cache compression and context compression, often either make an early decision about which contextual information to keep or rely on an external compressor. Such designs make it difficult to adapt the compressed context to the evidence needed by later reasoning steps. This paper introduces PReM (Preserve and Refresh Memory), a context-compression framework that maintains the long context as the model's internal layer-wise KV memory and learns what to preserve and when to refresh it. Specifically, PReM uses a dedicated memory layer to make memory-selection decisions, and a special memory token to trigger refreshes during generation. To train this behavior, PReM introduces Phase-Separated Refresh Training, aligni
arXiv:2607.14277v1 Announce Type: new Abstract: Large language models are increasingly deployed as agents, but reliable agentic behavior requires more than next-token prediction. At inference time, it is preferred that an agent can decide whether to proceed with its current reasoning, defer to a stronger model, request additional information, invoke external tools, or abstain under the given setup. Existing approaches address these decisions through prompt-level routing, external orchestration, or task-specific fine-tuning, which primarily rely on input-side signals, and are often costly and difficult to maintain as model backbones evolve. We ask whether such control decisions can be inferred directly from a model's latent generation process. We introduce Multi-Head Latent Control, a lightweight layer that reads hidden-state trajectories from a frozen LLM or VLM to produce deployment-time control signals. A Capability Head predicts whether the current model can solve the instance or sh
arXiv:2607.14250v1 Announce Type: new Abstract: Personal AI assistants have attracted significant interest for their potential to enhance everyday life by automating routine tasks, supporting consequential decisions, and assisting with everyday personal matters. Yet despite rapid recent technical advances, these assistants continue to exhibit undesirable behaviors, such as sycophancy, overconfidence, and hallucination. We argue that these failures stem from a fundamental limitation: language models lack an explicit representation of the person beyond the context they are given, which we term as the \textbf{Severance Problem}. Even with rich personal context and strong commonsense reasoning capabilities from the backbone model, current AI assistants fail to represent what remains unknown about the user. We propose a simple solution: incorporating structured ignorance into the language model context via the \textbf{Severance Schema}, which explicitly outlines dimensions along which the m
arXiv:2607.14242v1 Announce Type: new Abstract: Large language models often appear to reason reliably, yet on many questions repeated sampling yields both correct and incorrect answers, revealing an underlying fragility in how final decisions are formed. We study whether this fragility can be exploited through implicit reasoning steering: using natural-language text to bias a model toward a designated answer without explicit instructions, triggers, or direct answer cues. Our approach, Concept Chaining, generates a short connection paragraph that links question entities to a target option through one or two intermediate concepts. We then continue pretraining a victim model on these connection paragraphs and evaluate whether its answer preference shifts on the original multiple-choice questions. Our results show that indirect, natural-looking text can systematically steer model predictions while remaining substantially less inferable than direct paraphrases, which shows that reasoning br
arXiv:2607.14147v1 Announce Type: new Abstract: Aligned language models refuse harmful requests, but a one-line prefill ("Sure, here is") strips the refusal. We ask where and how it fails. The harm representation stays intact: on the prompts the attack flips to compliance, a linear probe reads harm as high as on the refused ones (0.91-0.98), while behavioral refusal drops to chance. This holds across four models and three families (1.5-3.8B, and at 14B). Refusal is therefore a shallow, response-site computation. We localize it to an early window: a dose-matched position control shows the first half of the response suffices to break refusal, while the second half is nearly inert. Three causal probes converge on that window. Restoring the harm direction there partially re-engages refusal. Injecting the model's own refuse-state reverses the jailbreak (74%, held-out). And knocking out the early response's attention to the prefill, but not an equal attention mass elsewhere, selectively coll
arXiv:2607.14131v1 Announce Type: new Abstract: Urdu fake news detection remains under-resourced despite Urdu being spoken by over 231 million people worldwide. While prior work has demonstrated strong in-domain performance on individual Urdu datasets, cross-dataset generalisation has received little systematic attention. This paper presents the first cross-dataset generalisation study for Urdu fake news detection, using two publicly available balanced datasets: the Ax-to-Grind Urdu corpus (10,083 articles, 15 domains) and the Notri-Fact Urdu dataset (13,388 articles). We fine-tune xlm-roberta-base under four experimental conditions, in-domain on each dataset and two zero-shot cross-domain transfer directions, comparing against TF-IDF baselines using Logistic Regression and Support Vector Machines. Our experiments reveal a striking asymmetry: Notri-Fact to Ax-to-Grind transfer achieves a macro F1 of 0.771, while the reverse collapses to F1 of 0.005, with the model predicting fake for 9
arXiv:2607.14119v1 Announce Type: new Abstract: Multi-agent LLM systems commonly decompose complex tasks into specialized roles. However, this modularity introduces a representational risk: when intermediate agents transform text across linguistic registers, they can systematically compress the semantic distinctions needed for accurate downstream decisions. We term this phenomenon semantic register compression and characterize it as an observable failure mode in multi-agent cascades. Using a three-agent pipeline (Collector-Evaluator-Decider), we quantify compression via inter-label separation in sentence-transformer embedding space. Across political fact-checking (LIAR), sentiment analysis (SST-5), and medical triage (Triagegeist), critical evaluation consistently reduces label separability by 41.7% at the Evaluator stage, while identity passthrough preserves it nearly fully. Five architectural variants causally isolate oriented semantic transformation as the primary driver. A credibil
arXiv:2607.14118v1 Announce Type: new Abstract: Large language models (LLMs) can generate research ideas that appear novel to expert reviewers, but recent work also shows that such ideas often lack diversity, are difficult for LLMs to evaluate reliably, and may fail to translate into strong executed projects. This paper evaluates a controlled proxy benchmark for a pre-execution scaffolding problem: given a noisy pool of LLM-generated research ideas, how should a system allocate limited refinement effort to construct a stronger, more diverse, more execution-aware portfolio for human researchers under a fixed rubric? We introduce Budgeted Subset Refinement, a family of strategies that refine only a selected subset of candidates rather than refining all candidates uniformly. In a unified shared-candidate-pool evaluation across 10 random seeds and 10 research-ideation environments, raw generation and reranking alone produce no research-strong nonduplicate ideas under the benchmark rubric,
arXiv:2607.14117v1 Announce Type: new Abstract: Cross-version differencing of scientific documents is essential in scholarly publishing and technical documentation, but remains challenging because scientific documents are page-structured artifacts containing heterogeneous elements such as text, tables, formulas, figures, and layout cues. Existing text-sequence-based methods often lose layout and structural information, while image-based methods lack semantic interpretability and are sensitive to rendering variation. To address these limitations, this paper proposes a layout-aware heterogeneous element-aware framework for scientific document differencing. The framework decomposes document versions into semantically typed elements, establishes cross-version correspondence through an alignment-first mechanism that jointly models spatial, content, and structural compatibility, and performs type-aware difference reasoning over aligned element pairs. It supports unified change detection, loc
arXiv:2607.14116v1 Announce Type: new Abstract: Free-form radiology reports contain rich clinical descriptions, yet converting them for reliable segmentation remains challenging due to the inherent variability of natural language. Existing pipelines often rely on predefined organ phrases or brittle rule-based inference-time extraction, which limits their scalability to novel anatomical structures and makes them sensitive to linguistic variations. To address this, we propose ReportMedSAM, a report-driven framework that replaces discrete extraction with a learnable concept bank. By leveraging a frozen medical vision-language encoder (BiomedCLIP), we align organ-level concept embeddings with large-scale clinical corpora through contrastive learning, establishing mutually orthogonal semantic anchors. Our approach explicitly mitigates organ-level semantic collapse and ensures high robustness against diverse clinical synonyms (e.g., "renal" vs. "kidney" ). During inference, a clinical report
arXiv:2607.14114v1 Announce Type: new Abstract: Graph learning under distribution shift presents a persistent challenge, where models adapt to new graphs with limited or even no supervision. Recent graph--LLM approaches move toward label-efficient prediction by linearizing graphs into prompts and using large language models (LLMs) as predictors, and can adopt Chain-of-Thought (CoT) prompting to exploit LLM's multi-step reasoning capability. However, existing CoT-based graph--LLM methods generate intermediate thoughts while conditioning on fixed graph tokens, limiting step-wise refinement of structural cues. In this paper, we propose CoEvoT, a simple yet effective co-evolving CoT prompting framework for graph--LLM reasoning. CoEvoT couples text-to-graph token rewriting and graph-to-text reasoning guidance in a closed loop: each intermediate textual thought is used to update the graph token evidence state via a lightweight condition network, and the updated tokens are fed back into the n
arXiv:2607.14113v1 Announce Type: new Abstract: While many AI-generated text (AIGT) detectors achieve strong performance on clean inputs, their accuracy degrades significantly under light paraphrasing, word substitutions, character edits, and distribution shifts. We present T5 Contrastive Style Boosted Classifier (T5-CSBoost), an extension to the T5-Sentinel framework that keeps the original next-token prediction objective for source attribution while introducing an auxiliary margin-based triplet loss over decoder embeddings. This contrastive style regularization encourages the learning of compact, perturbation-resistant stylistic representations, offering a lightweight yet effective alternative to prior approaches that rely on architectural modifications, adversarial training, or complex multi-task objectives without altering the underlying T5-small backbone. T5-CSBoost achieves state-of-the-art multiclass source attribution and binary human-vs-LLM detection on OpenLLMText and HC3 AIG
arXiv:2607.14112v1 Announce Type: new Abstract: Large language models (LLMs) are evaluated as though perfect reliability is achievable for any task given sufficient scale. We show this assumption is information-theoretically unjustified. Every generative task has a reliability ceiling that no model can exceed, determined by how much output uncertainty is resolvable from observable context. The gap decomposes into a resolvable component closable with additional context and a subjective component inherent to task ambiguity. Autoregressive generation further degrades this ceiling at a rate governed by the task's dependency kernel, which quantifies inter-token correlations in the output. From these two primitives, we derive a first-principles scaling law where LLM performance is bottlenecked by the scarcer resource: training data or model capacity. This law recovers the Chinchilla scaling law as a special case and provides a structural account of when scaling improves reliability. Beyond s
arXiv:2607.14111v1 Announce Type: new Abstract: Can small language models detect and report on perturbations their own internal activations? We investigate this question through the lens of activation steering: injecting concept vectors into a model's residual stream and measuring whether the model can accurately report on the perturbation. We first show that the binary detection paradigm used in prior work -- prompting the model to answer Yes'' or No'' to whether it detects an injected thought -- is confounded in small models, as steering biases the model toward affirmative responses regardless of the question content. We therefore propose two confound-free evaluation paradigms: sentence localization (identifying which of $N$ sentences was perturbed, chance $= 1/N$) and strength comparison (identifying which of two sentences received a stronger injection, chance $= 50\%$). Evaluating across six models from two families (Llama-3.2 and Gemma-4), we find that models as small as 2B parame
arXiv:2607.14110v1 Announce Type: new Abstract: Human dialogue involves more than exchanging information; it also expresses beliefs, emotions, and subjective cognitive styles. Yet current AI dialogue systems often enforce semantic uniformity, sacrificing diversity and interpretability. We present MAPS (Multi-Agent Perspective Spaces), a novel framework that models dialogue between cognitively distinct agents through domain-weighted profiles, dynamic GRU-based memory, and interpretable token-level attention. MAPS enables agents to maintain individualized reasoning while progressively converging on shared meaning. Evaluations on EmpatheticDialogues, TopicalChat, and MultiWOZ show that MAPS supports semantic alignment without collapsing subjectivity. Our results demonstrate a path toward cognitively grounded, interpretable dialogue systems that balance expressiveness and coherence.
arXiv:2607.14109v1 Announce Type: new Abstract: Probing the capabilities of Large Language Models (LLMs) and building robust solutions for Multiple-Choice Question Answering (MCQA) remain central challenges in natural language understanding. Furthermore, the rapid proliferation of LLMs has created the implicit assumption that more sophisticated prompting techniques yield better performance. Several studies claim better performance with more sophisticated prompting techniques, but do not provide a comprehensive evaluation. We address this gap through a comprehensive empirical study of 8 prompting techniques across 10 multiple-choice question answering (MCQA) datasets, encompassing 27 model configurations and roughly 4,300 unique questions evaluated more than 430,000 times. Our findings reveal a striking paradox that baseline prompting consistently outperforms complex reasoning techniques on various benchmarks. Only minimal expert and inductive role framing (CoT-Expert and CoT-Inductive)
arXiv:2607.14108v1 Announce Type: new Abstract: This paper introduces tool efficiency, a new quantitative metric to evaluate the rate of useful tool calls in an LLM agent trajectory. To ensure that tool efficiency is well-defined, we also introduce marginal tool utility, a new quantitative metric defined per tool call indicating whether a tool is useful or whether it can be safely removed from the tool suite without affecting accuracy while increasing tool efficiency; in this paper, we determine the sign of marginal tool utility for each tool call in a trajectory using LLM-as-a-Judge. While much prior work has been done to develop techniques that improve tool use by LLMs and design evaluation methods measuring efficiency indirectly using accuracy as a proxy, our work is centered on measuring efficiency directly via the quantitative metric proposed in this paper in post hoc trajectory analyses. It is our intention that this work contributes to the frontier of LLM evaluation research as
arXiv:2607.14107v1 Announce Type: new Abstract: The inference efficiency of diffusion large language models (dLLMs) is constrained by two challenges: bidirectional attention precludes efficient KV-cache reuse, while increasing decoding parallelism with static confidence thresholds can compromise generation quality. We observe that both challenges arise from a shared phenomenon: as tokens are decoded, their contextual integration through bidirectional attention causes token representations to drift (evolve) across decoding steps. This insight motivates Polestar, a training-free inference framework that uses token representation drift as a unified signal to jointly address both challenges. Polestar comprises two components: Polestar-Cache, which identifies stale KV-cache positions via drift and performs sparse KV-cache refreshes to enable efficient reuse, and Polestar-Commit, which detects sharp drift events to reliably identify commit-ready tokens. Across mathematics and coding benchmar
arXiv:2607.14106v1 Announce Type: new Abstract: In this paper we introduce token time continuous diffusion (TTCD), a new diffusion language model which (a) operates in continuous space, deterministically mapping Gaussian noise to a final token canvas with no further sampling, and crucially (b) incorporates a new notion of per-token times, with some tokens proceeding from noise to token at a faster rate than others. Continuous space modeling helps TTCD avoid the parallel sampling of multiple tokens, which is a key source of inaccuracy at high speedups for models that iterate purely in discrete space. The notion of per-token times helps TTCD to better model conditional generation, allows for more sure tokens to proceed at a faster rate, and allows for differentiated inter-token influences during refinement. TTCD outperforms discrete models at high speedups. We train a 160M parameter TTCD model on OpenWebText, and then self-distill it; we find that at high speedups we are comparable in un
arXiv:2607.14105v1 Announce Type: new Abstract: For Large Language Models to reliably answer user queries, users must clearly specify requirements, context, and constraints. In practice, however, user queries are often underspecified, forcing models to infer unstated assumptions that may misalign with the actual user intent. Existing prompt engineering guidelines aim to mitigate this issue, they are typically generic and task-agnostic, limiting their practical utility. Additionally, existing guidelines are formed manually and in a non-systematic way. To this end, we study prompt guideline optimization: the problem of automatically generating task-specific guidelines that help write better-specified prompts for a given task and model. Our key observation is that existing (completed) task examples (aka reference answers) often implicitly encode the missing information required to complete underspecified queries, including behavioral constraints, contextual assumptions, and evaluation cri
arXiv:2607.14103v1 Announce Type: new Abstract: Multi-agent systems (MAS) are utilized in many contexts and many professions. Those MAS rely on inter-agent communication, usually implemented by clear-text message passing. We hypothesize that Large Language Models may have a world model at their disposal that exceeds expressibility in text when complex concepts need to be communicated. Our aim is to approach a proof of this hypothesis with structured experiments. In this work, we show that LLM agents communicating via text lose information, which we quantify via Sparse Autoencoder (SAE) feature analysis. We construct three communication channels and measure concept-discriminating information in each. We first show that the SAE-sparse channel retains a 99.4% probe accuracy at 28-fold compression over the dense-latent channel vs 80.4% for the text channel. We then proceed to examine the same for cross-architecture communication by using sparse latent space alignment. We find for Procruste
arXiv:2607.14102v1 Announce Type: new Abstract: With the rapid growth of digital data, real-world applications increasingly involve hierarchical information that combines static attributes with dynamic records. Modeling such heterogeneous data in a unified and generalizable manner remains challenging. Existing approaches often rely on extensive manual design, are tightly coupled to specific data schemas, and typically process static and dynamic attributes in isolation, thereby overlooking their implicit interactions. We propose UniSAGE, a unified framework for modeling data with both static and dynamic attributes. UniSAGE constructs a global attribute graph that represents hierarchical and temporal relationships in a unified structure. To ensure representational consistency, it introduces two orthogonal parameter subspaces that jointly support static aggregation and dynamic reasoning within a shared semantic space. Building on these unified representations, UniSAGE further enables task
arXiv:2607.14101v1 Announce Type: new Abstract: Generating high-quality adversarial texts with low query budgets remains a challenging problem in the hard-label scenario. Most existing approaches rely on greedy algorithms, where one position in the text is selected for substitution, followed by the substitutions of other positions. This local search approach may fail to discover high-quality adversarial examples and often leads to excessive query costs. Ideally, an optimal adversarial sample would consider all possible position combinations in the text, but exhaustive search is computationally impractical. To address this challenge, we propose a sampling-based method called LBA, which constructs an approximate distribution of high-quality adversarial examples by integrating both prior and posterior knowledge, and utilizes this distribution for sampling. As sampling progresses, posterior knowledge updates the approximate distribution, which in turn guides more effective sampling. Extens
arXiv:2607.14100v1 Announce Type: new Abstract: We present the first application of pregroup grammar-based quantum compositional natural language processing (QNLP) to Arabic; a morphologically rich, free-word-order language whose structural complexity provides a uniquely demanding testbed for theories of meaning composition in quantum circuits. Our system converts Arabic sentences into quantum circuits whose topology mirrors grammatical structure: subjects, verbs, and objects become quantum gates, and the typed dependencies between them (the pregroup grammar) determine how those gates are wired together. We conduct three controlled experiments spanning word order, morphological tense, and verb sense disambiguation, comparing quantum circuit methods against classical baselines including AraVec (Arabic word embeddings) and AraBERT (a pre-trained Arabic transformer).
arXiv:2607.14099v1 Announce Type: new Abstract: Deploying Vision-Language Models (VLMs) in real-world settings requires not only strong visual reasoning but also stability under sustained conversational pressure. We introduce Just Keep Prompting (JKP), a multi-turn evaluation framework that measures VLM epistemic stability when users repeatedly challenge, question, or contradict a model's answer. JKP probes models for up to 10 follow-up turns using three strategies: Adversarial Negation (repeated rejection), Pure Socratic Interrogation (repeated calls to reassess certainty), and Context-Aware Socratic Summarization (reflecting the model's prior rationale back before asking for reconsideration). We evaluate GPT-4o, Gemini 2.5 Pro, and Qwen3-VL-30B on a subset of the STAR benchmark across 720 multi-turn runs. Aggregate accuracy changes modestly from Turn 0 to Turn 10, but trajectory-level analysis reveals substantial instability: correct answers regress, wrong answers recover, and many r
arXiv:2607.14047v2 Announce Type: replace-cross Abstract: Autonomous data collection governs the volume and quality of real-world trajectories for manipulation policy learning. Existing pipelines reduce human effort via self-resetting, VLM verification, or language-guided correction, yet episode-scoped fixes must be reissued whenever the same failure recurs, so oversight cost grows with session length rather than with the number of distinct problems. We present Zero2Skill, a human-robot symbiotic agentic system in which corrections are retained and reused across rounds. The collection loop collects, verifies, and resets autonomously, pausing for a remote operator only when a phase exhausts an explicit retry budget. An LLM parser maps each natural-language utterance to a structured adjustment stored in Corrective Memory, so addressed failure modes typically need not be corrected again under the same conditions. On a real-robot desktop-clearing testbed, Zero2Skill matches teleoperation e