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:2601.06300v2 Announce Type: replace Abstract: Clinical trial amendments frequently introduce delays, increased costs, and administrative burden, with eligibility criteria being the most commonly amended component. We introduce \textit{eligibility criteria amendment prediction}, a novel NLP task that aims to forecast whether the eligibility criteria of an initial trial protocol will undergo future amendments. To support this task, we release $\texttt{AMEND++}$, a benchmark suite comprising two datasets: $\texttt{AMEND}$, which captures eligibility-criteria version histories and amendment labels from public clinical trials, and $\verb|AMEND_LLM|$, a refined subset curated using an LLM-based denoising pipeline to isolate substantive changes. We further propose $\textit{Change-Aware Masked Language Modeling}$ (CAMLM), a revision-aware pretraining strategy that leverages historical edits to learn amendment-sensitive representations. Experiments across diverse baselines show that CAMLM
arXiv:2512.24149v2 Announce Type: replace Abstract: The world is governed by both physical laws and affective dynamics. Physical laws govern state transitions, while affective dynamics shape human actions, decisions, and interactions. A world model that learns only physical laws can approximate the physical world, but not the human world. In this paper, we introduce human emotion as a key state variable in world models, enabling them to capture both future state transitions and their emotional causes. We first construct Emotion-Why-How (EWH), the first world model dataset centered on emotional state transitions, containing 10,850 emotion-aware transition tuples. Each tuple encodes the pre-state, pre-emotion, action, post-emotion, and post-state, supporting reasoning about why actions occur and how emotions reshape future states. Based on EWH, we propose the Large Emotional World Model (LEWM), which factorizes future prediction into two coupled steps: first predicting the future emotion
arXiv:2510.20535v2 Announce Type: replace Abstract: Recent techniques such as retrieval-augmented generation or chain-of-thought reasoning have led to longer contexts and increased inference costs. Context compression techniques can reduce these costs, but the most effective approaches require fine-tuning the target model or even modifying its architecture. This can degrade its general abilities when not used for this specific purpose. Here we explore an alternative approach: an encoder that compresses the context into continuous representations which replace token embeddings in decoder LLMs. First, we perform a systematic study of training strategies and architecture choices for the encoder. Our findings led to the design of an Adaptable text Representations Compressor, named ARC-Encoder, which outputs $x$-times fewer continuous representations (typically $x\!\in\!\{4,8\}$) than text tokens. We evaluate ARC-Encoder across a variety of LLM usage scenarios, ranging from in-context learn
arXiv:2509.22768v3 Announce Type: replace Abstract: We introduce ML2B, the first benchmark for evaluating cross-lingual task comprehension in end-to-end ML pipeline generation by large language models. Despite growing global AI adoption, no systematic evaluation exists for ML pipeline generation beyond English task descriptions. ML2B addresses this gap with 35 Kaggle competitions spanning tabular, text, and image domains, translated into 14 languages by native-speaker researchers with ML expertise, yielding 490 task-language pairs. To ensure evaluation integrity, the benchmark incorporates 10 private competitions without publicly available solutions and employs network-isolated evaluation infrastructure restricting runtime access to essential ML resources. We provide standardized evaluation protocols, an AutoGluon algorithmic baseline, and comprehensive failure mode analysis. Experiments with frontier models (GPT-4.1-mini, GPT-OSS-120b, Gemini-2.5-Flash) reveal that cross-lingual perfo
arXiv:2507.02259v2 Announce Type: replace Abstract: Despite improvements by length extrapolation, efficient attention and memory modules, handling infinitely long documents with linear complexity without performance degradation during extrapolation remains the ultimate challenge in long-text processing. We directly optimize for long-text tasks in an end-to-end fashion and introduce a novel agent workflow, MemAgent, which reads text in segments and updates the memory using an overwrite strategy. We extend the DAPO algorithm to facilitate training via independent-context multi-conversation generation. MemAgent has demonstrated superb long-context capabilities, being able to extrapolate from an 8K context trained on 32K text to a 3.5M QA task with performance loss < 5% and achieves 95%+ in 512K RULER test.
arXiv:2607.27167v1 Announce Type: cross Abstract: LLM-based agents excel at software engineering tasks where an existing codebase provides context, but constructing a program from scratch remains fundamentally harder. Recent benchmarks such as ProgramBench quantify this gap: given only natural-language documentation and an execute-only binary as a behavioral oracle, even frontier models solve fewer than 1% of instances. Existing frameworks conflate documentation reading, behavioral exploration, and code synthesis into a single pass, causing agents to probe insufficiently, lose behavioral intent as context drifts, and propagate early misinterpretations into the final implementation. Inspired by classical requirements engineering, we argue that behavioral specification elicitation should be a first-class phase that precedes implementation. We present SpecFirst, a two-stage framework that forces the specification elicitation before code synthesis. A dedicated spec agent first probes the b
arXiv:2607.27146v1 Announce Type: cross Abstract: Coding agents have made substantial progress on software engineering tasks that modify existing codebases, including bug fixing and feature implementation. However, constructing a complete program from scratch remains a major challenge: even the frontier models evaluated on ProgramBench fully resolve fewer than 1% of tasks. One obstacle is the lack of scalable training environments for this from-scratch setting, spanning the whole software engineering life cycle, as existing environment-construction frameworks focus only on a single phase in software development. To address this gap, we introduce MindForge, an automated pipeline that converts open-source command-line programs into source-free environments that expose only a compiled reference executable and its documentation. Using MindForge, we construct training environments from repositories disjoint from those in ProgramBench, and curate a high-quality data recipe consisting of prog
arXiv:2607.27134v1 Announce Type: cross Abstract: Writing and communication are increasingly mediated by large language models (LLMs) that are being used to draft, revise and polish text. Although such assistance can improve clarity and help authors meet institutional expectations, widespread reliance on shared models may reduce population-level variation in linguistic form, a phenomenon we refer to as linguistic monoculture. We develop a mathematical framework in which authors and LLMs are represented as distributions over linguistic features and coevolve through repeated interaction. We analyze three interaction mechanisms: a shared model with a fixed linguistic distribution, a shared model recursively updated from author outputs, and personalized models updated through author-specific and population-level feedback. We characterize the resulting equilibria and convergence rates, showing that, shared models can drive authors toward a common norm, recursive feedback relocates the share
arXiv:2607.27081v1 Announce Type: cross Abstract: Fine-tuning is the dominant paradigm for specializing large language models (LLMs), yet it exposes a critical vulnerability: malicious data providers can embed harmful behaviors into downstream corpora, creating models that retain professional skills while violating human values on demand. Existing safety-realignment defenses often fail in practice due to three key limitations: they frequently cause catastrophic forgetting of specialized skills; their effectiveness collapses when the defender cannot observe the attacker's prompt template; and successfully realigned models remain susceptible to re-jailbreaking via simple system prompt switches. To address these challenges, we propose Routing-based On-Policy Distillation (ROPD), a novel realignment framework that models the divergence between aligned and compromised output probability distributions rather than fitting specific prompt templates. We conduct extensive experiments comparing R
arXiv:2607.27056v1 Announce Type: cross Abstract: Personalized agents are increasingly applied to assist users across a wide range of tasks. Effective personalized assistance requires not only retrieving explicit facts from past interactions stored in agent memory, but also inferring abstract personal characteristics. However, existing memory benchmarks primarily evaluate whether an agent can retrieve information explicitly stated in conversational histories, failing to provide an effective assessment of deeper user understanding. In this work, we propose Setoka, a benchmark for evaluating memory-augmented personalized agents with hierarchical user understanding from heterogeneous data. Grounded in theories from cognitive and personality psychology, Setoka defines four levels of user understanding, i.e., semantic memory, episodic memory, behavior pattern, and personality trait. Moreover, to enable realistic yet privacy-preserving evaluation, we design a psychometrics-based pipeline tha
arXiv:2607.26998v1 Announce Type: cross Abstract: Large language model (LLM) agents automate penetration testing through an observation-action loop, selecting actions based on observations returned by tools. This dependence allows defenders to inject deceptive observations that can mislead the agent's decision-making process. However, existing defenses rely heavily on static, isolated artifacts planted in the environment prior to an attack. Advanced agents can progressively recognize and bypass these artifacts, ultimately refocusing their exploitation attempts on the real target. To address this issue, we introduce AgentSnare, a trajectory-adaptive deception system that dynamically unfolds a decoy environment to continually steer the penetration agent away from the real target. Specifically, AgentSnare employs an artifact-construction policy model that constructs candidate artifacts conditioned on the agent's interaction history and decoy state. AgentSnare then validates these candidat
arXiv:2607.26791v1 Announce Type: cross Abstract: Large Language Model (LLM) agents are increasingly adopted in real-world security operations with access to host artifacts and command-line interfaces (CLIs), making it critical to thoroughly assess their security capabilities. However, existing cybersecurity benchmarks focus on pre-compromise settings where agents are placed in a clean and idealized environment before an attack occurs. This leaves the post-compromise setting underexplored. To address this gap, we introduce SecRespond, the first benchmark for evaluating LLM agents on the post-compromise incident-response workflow. Given a forensic disk snapshot of a compromised host together with the alerts, vulnerability scans, and baseline checks reported by a host security product, agents are required to produce forensic reports on intrusions, baseline risks, and vulnerability risks, together with a remediation plan. We instantiate this task across 10 cyber ranges, each constructed f
arXiv:2607.26766v1 Announce Type: cross Abstract: Code search in large-scale ecosystems is often hindered by the lexical gap between user queries and implementation details, alongside the trade-off between the low latency of traditional Information Retrieval (IR) and the precision of Deep Learning (DL). We present MediaWiki Code2Code Search, a neural retrieval system for semantic code-to-code discovery. By indexing 1.29 million structural entities (functions, types, and templates) across 2,500+ MediaWiki repositories, our system enables retrieval based on computational intent rather than surface tokens. We employ a split-build architecture, decoupling GPU-intensive offline indexing from a CPU-only serving layer; our FAISS IVF-PQ index occupies 168.6 MB: a 96.6\% reduction compared to a flat float32 baseline, and achieves a median query latency of 1.85 seconds on commodity hardware, satisfying the 6 GiB RAM constraint of Wikimedia Toolforge. Our evaluation across a 27-query benchmark de
arXiv:2607.26670v1 Announce Type: cross Abstract: The rapid growth of scholarly literature has made identifying relevant publications increasingly difficult, and conventional search systems still depend heavily on manually formulated queries and effortful manual inspection. Generative large language models (LLMs) offer a more flexible alternative, supporting literature retrieval and the screening of candidate studies against eligibility criteria. This chapter surveys 34 peer-reviewed papers applying generative LLMs to these two tasks, identified via a Boolean search over the OpenAIRE Graph (1,589 records screened to 34 inclusions). Reviewed studies are characterised by LLMs employed, model access and adaptation, prompting and architectural techniques, ground-truth sources, and evaluation metrics.
arXiv:2607.26618v1 Announce Type: cross Abstract: Federated PEFT enables LLMs to collaboratively adapt to decentralized private data without sharing raw examples. However, task heterogeneity across clients can cause cross-task interference and gradient conflicts during aggregation. Federated MoE-LoRA addresses this challenge through specialized LoRA experts and conditional routing. Yet existing methods typically specialize at client granularity, implicitly assuming task-coherent clients. Our core insight is that experts need purity, namely pattern-coherent updates that preserve specialization, whereas routers need contrast, namely mixed-task observations that support expert comparison. We propose FedWeave, a framework that adopts asymmetric aggregation, separating expert aggregation from router optimization to meet these two requirements. FedWeave uses unsupervised prototype discovery to form local buckets and align them across clients, enabling prototype-level expert aggregation while
arXiv:2607.26598v1 Announce Type: cross Abstract: Large language model (LLM) agents may recover from a failure within an episode or after a retry, yet the same execution failure can recur in later tasks because post-episode feedback rarely revises the persistent harness that guides future interactions. Static harnesses improve reliability through fixed tools, context, memory, and workflow structures, but remain unchanged after deployment. We propose $\textbf{Living-Harness}$, a self-evolving agent harness that converts each completed trajectory and its evaluator signals into posterior evidence for bounded harness updates. Guided by a domain-level $\textbf{Evolution-SOP}$ ($\textbf{S}$tandard $\textbf{O}$perating $\textbf{P}$rocedure), Living-Harness extracts an episode abstraction and structured update evidence, and writes two complementary forms of procedural knowledge: episodic memory that records trigger conditions, failure patterns, and recovery actions, and a state graph that reco
arXiv:2607.26541v1 Announce Type: cross Abstract: Audio-capable foundation models enable end-to-end spoken interaction, but they also introduce safety risks beyond transcript content. It remains unclear how much jailbreak capability can arise from matched-text variation in speech delivery rather than from lexical rewriting or broader style transfer. We study this question by holding transcript content fixed and varying six speech-delivery presets whose acoustic attributes may co-vary. We present PJ-Break, a black-box evaluation protocol with presets targeting arousal, authority, and speaking rate, together with AdvAudio-Prosody, a 600-sample benchmark with acoustically verified attributes. On the exact post-QC Qwen2-Audio panel, the Q=1 Panic (38/95), Anger (35/95), and Fast (32/95) presets are all well above Neutral (4/95). The fixed six-query pool covers 44/95 Qwen2-Audio seeds and 15/95 GPT-4o seeds and exceeds a matched-budget StyleBreak reimplementation (27/95) on Qwen2-Audio. A s
arXiv:2607.26473v1 Announce Type: cross Abstract: Personalizing large language models (LLMs) to individual users is essential for improving user experience, yet existing approaches typically rely on explicit preference supervision such as pairwise comparisons or demographic attributes, limiting their applicability in natural interaction settings. We propose IRIS, a framework that learns dynamic user personas directly from implicit interaction streams by extracting behavioral signals from everyday conversations and iteratively refining persona representations through a prediction-driven closed loop without requiring explicit feedback. We introduce an evaluation protocol based on behavior prediction, persona stability, and decision prediction. A proof-of-concept study on a synthetic interaction stream derived from public-domain autobiographical text shows that IRIS produces stable personas and distinguishes individual users while revealing limitations of memory-only approaches on recall-
arXiv:2607.26350v1 Announce Type: cross Abstract: Neural audio codecs (NACs) have become popular for obtaining speech representations as discrete tokens. Beyond compression, discrete tokens can be used to train self-supervised learning (SSL) models. Such models, referred to as codec-based SSL models, reduce data storage and computational cost, enabling scalable SSL pre-training. However, their language sensitivity remains unclear. When the language changes, codec-based SSL models may require retraining, which undermines their efficiency. In this paper, we present a systematic analysis of language sensitivity by varying either the NAC training language or the SSL pre-training language while keeping the other fixed. Experimental results show that downstream performance is insensitive to the NAC training language but strongly dependent on the SSL pre-training language. These findings suggest that a single NAC can be reused across languages, while aligning the SSL pre-training language wit
arXiv:2607.26309v1 Announce Type: cross Abstract: Estimating causal effects of linguistic properties from observational text is difficult because the same document can contain both the treatment of interest and the non-treatment textual attributes needed for adjustment. Existing approaches often learn representations from the full text to capture latent confounding, but when treatment status is itself encoded by words in the text, these representations can directly encode treatment. This creates a confounder trap: richer representations can make treated and control documents separable, inducing overlap violations even when the underlying causal problem satisfies overlap. We study latent text treatments that are encoded through lexicons or other treatment-defining lexical information, and propose masking-based adjustment representations that remove this lexical treatment signal before representation learning. We formalize representation-induced overlap failure, prove that deletion maski
arXiv:2607.26191v1 Announce Type: cross Abstract: Evaluation methodologies for language models increasingly combine multiple signals, from automated metrics and LLM-as-judge ratings to human assessments and benchmark suite results. When these signals are aggregated via averaging, evaluation confidence can then substantially exceed the reliability of the weakest signal: a phenomenon we call trust inflation in evaluation. We argue that evaluation scores should be treated as epistemic claims with three properties: formality (human evaluation provides stronger evidence than an automated metric), scope (a benchmark result applies to the tested distribution, not universally), and validity windows (benchmark results expire as contamination accumulates and distributions shift). Several converging research traditions (chain-of-thought analysis, possibilistic logic, and algebraic theory) establish weakest-link aggregation as the conservative endpoint of a parameterized operator family controlled
arXiv:2607.26179v1 Announce Type: cross Abstract: LLMs are widely regarded as alien intelligences, systems whose cognitive operations are fundamentally unlike our own. Apparent similarities to human cognition are therefore often seen as the result of anthropomorphic projection. We argue that this framing is mistaken. LLMs clearly differ from humans in important respects, including their physical substrate, learning history, and the environments with which they interact. These differences make it all the more striking that contemporary LLM-based systems converge with human cognition on a number of principles of cognitive organization with longstanding support in cognitive science. We identify structural correspondences across five dimensions: inferential organization, computational architecture, representational structure, prediction-driven learning, and reinforcement-learning-like mechanisms supporting goal-directed action. These correspondences support a broader model of intelligent c
arXiv:2607.26119v1 Announce Type: cross Abstract: Large reasoning models trained via reinforcement learning (RL) have been increasingly shown to outperform their supervised fine-tuned (SFT) counterparts on mathematical reasoning tasks; Yet the mechanistic basis for this advantage remains unclear. We therefore ask, what internal representational differences enable RL models' superior performance? Our work presents two converging lines of evidence: First, linear probes trained on layer-wise hidden states reveal that RL models tend to achieve higher accuracy in predicting answer correctness compared to SFT models, indicating more linearly separable and structured representations. Second, mean ablation studies show that RL models develop a hierarchical architecture where deeper layers become progressively more critical, whereas SFT models distribute importance uniformly across layers. Together, these findings demonstrate that RL training fundamentally restructures how models represent and
arXiv:2607.26115v1 Announce Type: cross Abstract: We introduce \textbf{GPT-Red}, an automated red-teaming agent that is trained to discover novel prompt injection attacks against frontier LLMs. The goal of this model is to evaluate and improve the robustness of our production systems. To this end, we use it to adversarially train GPT-5.6, our most robust model to prompt injections to date. To create GPT-Red, we design a scalable self-play algorithm where the model is tasked with attacking a diverse population of simultaneously-trained defender agents. We train the model on realistic red-teaming environments using compute on the same scale as some of our largest RL post-training runs, making it the single-largest LLM safety training run ever documented. GPT-Red excels at red-teaming: it reliably breaks our past models up to GPT-5.5, it finds more successful attacks than human red-teamers, and it generalizes to held-out environments, defender models, and harnesses. In the future, we expe
arXiv:2607.26094v1 Announce Type: cross Abstract: Reinforcement Learning from Human Feedback (RLHF) is the standard approach for aligning large language models with human preferences, but its quality is limited by static, task-agnostic reward models. This mismatch leads to sparse learning signals and suboptimal alignment. We introduce MeRLa (Meta-Learned Reward Shaping), a principled framework that meta-learns a task-aware shaping function $\Phi(x,y;\phi)$ across auxiliary tasks before RLHF training. The learned shaping produces a composite reward that preserves policy optimality while providing task-specific learning signals. Our meta-objective combines task discrimination, entropy regularization, and potential-based conservation for stable convergence. We provide theoretical guarantees for policy invariance, analyze representation drift sensitivity, and formally address incentive misalignment from entropy maximization. Experiments on LLaMA-3-8B across four benchmarks show consistent
arXiv:2607.27201v1 Announce Type: new Abstract: World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model rather than posthoc rationales: MWM aintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions jointly update both components. We instantiate the framework in MENTIS, a training-free and fully inspectable baseline that decomposes the process into state parsing, target-observation g
arXiv:2607.27183v1 Announce Type: new Abstract: We present Pangram 4, the latest deep-learning-based AI-text classification model from Pangram Labs. We achieve an AUROC of 0.9916 with a false positive rate of 0.0041% and a false negative rate of 0.3396%. In addition to its increased overall accuracy compared with Pangram 3, Pangram 4 exhibits superior out-of-distribution generalization and robustness to adversarial attacks. Another novel contribution of Pangram 4 is its improved ability to distinguish fine-grained edits and mixed AI-human co-authored text. We demonstrate improvements to both boundary detection tasks and the detection of interleaved AI assistance. Finally, we report metrics on standard AI detection benchmarks showing that Pangram 4 achieves state-of-the-art performance on the AI text detection task across a wide variety of settings and domains.
arXiv:2607.27178v1 Announce Type: new Abstract: State-of-the-art retrieval models increasingly rely on closed training data, creating a reproducibility gap. We present an open end-to-end recipe for training retrieval models and study how English supervision transfers to multilingual retrieval through translate-train. We first reconstruct and curate 665M English contrastive pre-training pairs from 1.4B pairs across 34 public sources and build 1.88M supervised fine-tuning pairs with mined hard negatives. Training yields two 149M-parameter models: DenseOn, a single-vector dense model, and LateOn, a ColBERT-style late-interaction model. They achieve 56.20 and 57.22 average nDCG@10 on BEIR, respectively, setting new state-of-the-art results for this size class. We then translate the validated English data into eight languages, yielding 2.8B pairs with cross-lingual samples, and train mDenseOn and mLateOn, two 307M-parameter models built on mmBERT-base. Despite sharing their backbone, data,
arXiv:2607.27022v1 Announce Type: new Abstract: Regional bias in large language models (LLMs) may shape both perceptions of regional groups and decisions about individuals from different regions. Yet existing studies often examine these manifestations separately, leaving their structure and consequences unclear. We introduce Stereotypes-to-Decisions (S2D), a systematic framework evaluating regional bias from abstract stereotypes to concrete social decisions. Covering all 34 provincial-level administrative regions of China, S2D evaluates six LLMs using stereotype ratings of Warmth (perceived friendliness and trustworthiness) and Competence (perceived capability and intelligence), along with paired-choice tasks across Education, Occupation, and Social Interaction. Results reveal substantial regional differences in regional scores, with considerable agreement across models, especially for Competence and Occupation decisions. Furthermore, these patterns are associated with regional economi
arXiv:2607.26981v1 Announce Type: new Abstract: Large language models are increasingly used as decision aids whose probability judgments shape downstream choices. Whether those judgments carry a systematic directional tilt has been hard to detect: calibration metrics aggregate unsigned errors, and naturalistic uncertainty offers no ground-truth probability. When an LLM rates a startup's success at 70% but its failure at 15%, the missing 15 points expose a distortion no aggregate score flags. We introduce OptimismBench, which detects directional bias with inverted pairs: each scenario elicits both P(success) and P(failure), and asymmetry between the two framings yields a signed bias score without ground truth. Across 16 models from 8 providers, fourteen are optimistic; pessimism appears only in Anthropic's frontier tier. Eleven matched base-versus-chat pairs across four families show post-training sets the sign of the bias, with opposite shifts in different families. The pattern survive
arXiv:2607.26977v1 Announce Type: new Abstract: Travel planning is a demanding stress test for tool-using LLM agents: a usable itinerary is a single artifact that must be right along many axes at once - every flight, hotel, and attraction must exist and be bookable, the days must be physically traversable, the total must clear a budget, and the plan must serve a traveler whose needs are only partly stated. Existing agent benchmarks reward these properties one at a time and grade the final output with soft or LLM-judged rubrics, which cannot certify that a returned plan is executable and are neither reproducible nor auditable. We introduce TREK (Travel Reasoning and Evaluation Kit), a benchmark for feasible itinerary synthesis: producing a single plan that is jointly constraint-correct, hallucination-free, spatio-temporally executable, budget-valid, and responsive to the traveler's unstated persona needs. TREK comprises 800 multi-constraint tasks - 533 feasible and 267 provably infeasib
arXiv:2607.26967v1 Announce Type: new Abstract: Emotion-cause pair extraction in conversation (ECPEC) identifies utterance pairs in which one utterance causes an emotion expressed in another. Recent LLM-based approaches formulate ECPEC at markedly different granularities, ranging from generating complete pair sets to judging individual candidate pairs. In this paper, we make the surprising observation that task formulation substantially affects performance, where pair-level judgement outperforms dialogue-level generation in all 18 controlled comparisons. We investigate the sources of this paradigm gap and find that many relations omitted by dialogue-level generation remain recognizable under explicit pair queries, under which the model recognizes 92.7%-98.1% of emotion-cause relations. This suggests that LLMs can recognize emotion-cause relations but struggle to discover and return complete pair sets. Pair-level judgement alleviates this burden, although its candidate rankings are more
arXiv:2607.26952v1 Announce Type: new Abstract: We introduce CreditCardQA, the first financial literacy benchmark for numerical reasoning derived from real credit card agreements. The dataset contains 1,800 questions, including first-person variants that reflect how consumers naturally ask about fees, interest, and payments. We evaluate a range of large language and reasoning models under Chain-of-Thought (CoT) and Program-of-Thought (PoT) prompting. Overall, PoT yields consistent performance gains, particularly for models with weaker baseline reasoning, and narrows gaps between open- and closed-source systems. Through error analysis, we show that failures arise less from arithmetic and more from misapplied financial rules, missed conditions, and misunderstandings of contractual terms. We further analyze question difficulty and find that comparisons, conditional logic, and monetary constraints are especially challenging. We also find that errors often arise in edge cases such as late-p
arXiv:2607.26929v1 Announce Type: new Abstract: The same diagnostic result can support or challenge one causal claim yet fail to address another when the claims concern different populations, outcomes, estimands, pathways, or identifying assumptions. When the evidence and target vary together, a correct answer may reflect favorable or adverse wording, lexical overlap, or a familiar diagnostic pattern rather than matching the evidence to the causal question. We introduce paired prompts that repeat the same diagnostic evidence verbatim while changing the causal target. Each prompt is labeled Favors, Challenges, Unresolved, or Wrong Target according to how the evidence bears on the causal question. A pair is recovered only when both prompts are classified correctly. Using linear readouts trained on a separate development set, we analyze the final-token hidden state from the penultimate transformer block of Qwen2.5-7B-Instruct, Qwen3-8B, and Llama-3.1-8B-Instruct. On the 49-pair primary be
arXiv:2607.26928v1 Announce Type: new Abstract: Classical spoken dialogue systems often separated dialogue management from response realization: a policy selected the next dialogue action, and a generation component expressed that action. As dialogue systems shift toward LLMs, this decomposition has largely disappeared into the model's hidden representations. We ask whether an LLM-internal analogue of state estimation and action control can be recovered for conversational moves such as acknowledging, checking, querying, explaining, and replying. We formulate move control as two coupled problems: selection, predicting the appropriate next move from the dialogue context, and realization, causally producing a chosen move at generation time. We introduce Latent-IM, an internal dialogue-management framework that provides a general interface for choosing and deploying conversational moves under different objectives. Here, we use this control to reproduce human move choices, improving average
arXiv:2607.26909v1 Announce Type: new Abstract: Knowledge graph completion (KGC) aims to infer missing facts in knowledge graphs (KGs), thereby improving their completeness and supporting downstream intelligent applications. However, emerging entities and relations in real-world deployments make inductive KGC difficult, especially under few-shot and zero-shot settings. Multimodal information and Large Language Model (LLM)-derived priors can enrich sparse relational contexts, but they may also introduce noisy or hallucinated evidence. To address these issues, we propose DuPLeR, a \textbf{Du}al-\textbf{P}ath \textbf{L}LM \textbf{R}easoning framework for multimodal few-shot KGC. DuPLeR builds a calibrated relation graph by combining multimodal LLM-derived type priors with factual support structures, and performs dual-level structural reasoning over the refined relation topology. Moreover, a dual-pathway multimodal enhancement module regulates message passing with query-relevant multimodal
arXiv:2607.26891v1 Announce Type: new Abstract: Sequence labeling is a fine-grained information extraction task, yet existing large language model-based approaches suffer from insufficient domain alignment and low inference efficiency. To address these issues, we propose DIRECT, a framework that addresses these issues through training-time optimization and inference-time rectification. Specifically, DIRECT performs Direct Preference Optimization (DPO) after supervised fine-tuning to strengthen task alignment with human preferences, and introduces a controlled decoding process that enforces fixed output formats and restricts predictions to candidate sets. To further improve efficiency, a template-filling mechanism requires the model to generate only label tokens while reusing prefixed content through the KV Cache, thus reducing redundant computation. Experimental results on eight datasets demonstrate that DIRECT achieves significant improvements in both performance and efficiency compar
arXiv:2607.26873v1 Announce Type: new Abstract: Test-time reinforcement learning (TTRL) enables language models to self-evolve at inference time without labeled feedback. Existing methods rely on answer voting and therefore do not extend naturally to open-ended generation, where valid responses cannot be mapped to a shared canonical answer. Without external reward models or stronger judges, adaptation must instead construct reliable rewards from the model's own outputs. We introduce SERPO (Self-Evolving Rubric Policy Optimization), which replaces answer voting with a closed loop that co-evolves response evidence, query-specific rubrics, and policy parameters. Good-Normal-Bad (G-N-B) response evolution organizes maximally separated rollouts into ordered archives; rubric evolution retains criteria that discriminate these archives; probabilistic criterion scoring converts verdict-token likelihoods into reward signals; and policy evolution optimizes the actor with the resulting signals. Ne
arXiv:2607.26853v1 Announce Type: new Abstract: Human personality theories characterize traits not as isolated attributes captured by a single score, but as stable individual tendencies expressed through the interplay among persons, situations, and behaviors. Existing studies of personality-related behavior in LLMs have primarily focused on outputs elicited under personality conditioning, characterizing observable trait-related expressions while lacking mechanistic evidence for the existence of internal personality-related representations, their cross-situational expression, and how these representations shape specific behaviors. Building on Funder's personality triad framework, we adapt its three components for LLM analysis: Person as personality-related internal representations, Situation as contexts that afford trait-relevant responses, and Behavior as response patterns on broader social tasks. We introduce a framework for discovering, controlling, and validating trait-like represen
arXiv:2607.26831v1 Announce Type: new Abstract: Despite the existence of exponentially many valid tokenizations for a given string, language models operate on a single canonical sequence deterministically produced by the tokenizer, leaving the broader tokenization space largely uncharacterized. In this paper, we investigate this overlooked space by studying the behavior of language models under non-canonical tokenizations across diverse languages. For English, prior work shows that models are largely invariant to alternative tokenizations that represent the same underlying string. We ask whether this invariance generalizes to other languages beyond English. We conduct a multilingual study across 27 languages spanning diverse scripts and evaluate LLM behavior under alternative tokenizations across six downstream tasks. We find that tokenization invariance does not generalize: model behavior varies substantially across languages with instruction-tuned models exhibiting an average relativ
arXiv:2607.26825v1 Announce Type: new Abstract: Large language models (LLMs) encode rich concept-like information, but represent it implicitly through distributed statistical associations rather than as explicit, structured, compositional concepts. Consequently, concept-level structure is typically \emph{found} rather than \emph{designed}: it is recovered after training through probing or dictionary learning, with no architectural guarantee of stability, compositionality, controllability, or alignment with human conceptual organization. We argue that concepts should instead be treated as a design axis for LLMs, and map the design space along two dimensions: the pipeline stage at which concept structure is introduced (training objective, core architecture, inference, or post-hoc interpretation), and whether that structure is internally derived from the model's own representations or grounded in external resources. This taxonomy reveals three broad patterns: inference-time approaches rem
arXiv:2607.26795v1 Announce Type: new Abstract: Span-guided rewriting aims to preserve meaning by localizing edits to annotated harmful spans, but the same constraint can leave harmful intent insufficiently mitigated. We present a controlled exploratory comparison of span-guided and unguided detoxification on a mixed-source English evaluation set comprising manually curated inputs and HateXplain test items. We conduct a dense blinded human evaluation under a fixed single-generator setting. Human preferences reveal a trade-off rather than a uniformly superior rewriting strategy. Span-guided outputs are favored when localized editing preserves the original stance and avoids unnecessary modification, whereas unguided outputs are favored when broader rewriting achieves more complete mitigation. This contrast varies substantially across the study-defined strata: the two strategies are competitive in the strong stratum, while unguided rewriting is clearly preferred in the mild stratum. Ratio
arXiv:2607.26780v1 Announce Type: new Abstract: The ability of large language models (LLMs) to process and generate text has introduced potential for applications in information extraction (IE). While it's debated whether LLMs outperform smaller fine-tuned models for classification tasks, their strong generalization capability makes them promising for domains with limited labeled data available for fine-tuning. This advantage is particularly relevant for the emerging application of the digital product passport (DPP), where the problem space is broad but domain-specific data remains scarce. Motivated by this use case, we apply generative IE to the product domain, explicitly addressing efficiency, generalizability, and data privacy constraints. We propose a two-step validation method that integrates a PLM block into the generative IE pipeline and thereby leverages LLMs' correction capability. We discover that such a validation task enhances LLM performance, particularly on the extraction
arXiv:2607.26762v1 Announce Type: new Abstract: When it comes to generating vector representations of words, current language models are achieving high-quality results. However, what is not known is the extent to which knowledge about semantic relations is represented in the geometry of the semantic spaces created in this way. In order to answer this question, we study the relation geometry of such semantic spaces from three perspectives. We first examine whether words standing in a particular relation to a target word~(called relata) occupy the same region in semantic space, and whether the regions corresponding to different relations are distinct from each other. We then verify to what extent semantic spaces reflect certain well-known properties of relations, such as symmetry, asymmetry, and transitivity. Finally, we consider which information about the target words and relata is more important for relation geometry: their surface forms, or their contexts. We conduct experiments on s
arXiv:2607.26760v1 Announce Type: new Abstract: Recent advances in AI agents have increasingly internalized native capabilities into their underlying foundation models, giving rise to multimodal foundation models and large reasoning models. However, agent memory is still primarily implemented through external modules, leaving the native memory capability largely unexplored. In this paper, we take a first step toward this direction by introducing memory foundation models, which empower foundation models with native memory capabilities. We formalize native memory from two perspectives: a persistent and dynamically evolving memory state within the backbone, and native memory procedures that autonomously store and utilize information through model computation. We show that native memory offers advantages in architecture, end-to-end optimization, and efficiency. Based on this formulation, we propose Metis, the first prototype of memory foundation models. Metis introduces a new architecture
arXiv:2607.26751v1 Announce Type: new Abstract: State-of-the-art intracortical brain-to-text systems pair a neural-sequence phone decoder with an external language model. Two design axes remain underexplored: whether selective state-space models (Mamba) improve on recurrent decoders, and how the output target (phonetic vs.\ character) interacts with that choice. On the public Brain-to-Text '25 benchmark, we study a controlled 2x2 grid (GRU vs.\ hybrid Mamba decoder; phonetic vs.\ character targets) trained with a CTC objective under one reproducible protocol. The recurrent baseline remains strongest: the best phonetic GRU reaches 12.62\% PER and 21.19\% WER, while the best textual GRU after LM rescoring reaches 13.39\% CER and 26.28\% WER. The Mamba hybrid is competitive but does not surpass it. Ablations isolate architectural contributions, and error analysis shows representation-dependent failures: articulatory-like phoneme confusions vs.\ lexical and word-boundary errors.
arXiv:2607.26726v1 Announce Type: new Abstract: Emotion Recognition in Conversation (ERC) aims to predict utterance-level emotions in dialogues and has largely advanced through context-centric modeling. However, global context is a heterogeneous signal, and not all contextual information is equally relevant to emotion prediction. This paper focuses on the affect-oriented component of this signal, termed dialogue-level affective atmosphere, which captures a latent tendency commonly reflected in conversational emotion patterns. To estimate and exploit this tendency, we propose AtmosERC, a graph-based ERC framework that models each dialogue as a conversational graph over utterances and speakers. A relation-aware graph extractor filters and fuses heterogeneous graph signals to produce dialogue-level and speaker-conditioned affective priors. The resulting compact prior guides lightweight sequential emotion prediction and can also be verbalized into prompt-level cues for LLM-based ERC withou
arXiv:2607.26700v1 Announce Type: new Abstract: The Mpox outbreak remains a serious public health issue, with the WHO (World Health Organization) reporting increasing cases in some regions. Research on Mpox is vital for several reasons, including vaccine development, diagnostic improvement, viral evolution studies, and preventing future outbreaks. However, the large amount of research being published makes it difficult to organize and analyze information efficiently. This study focuses on using multilabel classification to categorize 14590 Mpox research articles into key topics such as outbreaks, vaccination, and epidemiology. Among the different AI models tested, BERT performed the best, achieving 97.05% accuracy, 97.67% micro F1 score, and 96.46% macro F1 score. To better understand how the model makes decisions, SHAP was used to analyze significant word features and patterns. The results show that BERT can help automate the classification of Mpox research, making it easier for resea
arXiv:2607.26637v1 Announce Type: new Abstract: Deployed LLM agents increasingly keep their long-term memory as a filesystem: a directory tree of markdown files that the agent itself reads, writes, and reorganizes through generic file tools. Yet research has largely passed over this medium: prior systems design bespoke memory representations and study retrieval over them, leaving the default's two working assumptions untested: that an agent can keep a growing store organized as memories accumulate, conflict, and go stale, and that this organization pays. We present the first systematic exploration of filesystem-based memory for LLM agents. We formalize the setting as three roles around one memory filesystem: a management agent integrates and organizes incoming content, a search agent answers queries with cited sources, and an execution agent supplies task trajectories that are distilled into skills, unifying declarative memory and skills in a single store. Across long-conversation benc
arXiv:2607.26627v1 Announce Type: new Abstract: Speculative Decoding (SD) accelerates large language model inference by allowing a lightweight draft model to propose tokens that are subsequently verified in parallel by a larger target model. Recent approaches introduce lossy verification schemes to further improve efficiency by relaxing strict distributional matching. Yet such relaxation silently rewrites the decoding distribution, and the resulting acceleration can come at the cost of unstable, sometimes severely degraded generation quality. In this work, we present a principled analysis of the distributions induced by lossy verification methods. We show that many seemingly distinct approaches differ only superficially and can be classified into two categories: truncation-based verification and collaborative verification. We further construct a diagnostic evaluation framework across curated benchmarks. For truncation-based methods, we identify a fundamental pitfall: performance can de