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.
Curative Insurance Company partnered with Wondr Health to expand weight management support for its members. The post Curative, Wondr Health Team Up for Weight and Metabolic Health Support appeared first on MedCity News .
Attovia Therapeutics’ biologic medicines could offer advantages over currently available immune disorder drugs, such as Sanofi’s blockbuster medicine Dupixent. The IPO comes as Attovia looks to expand clinical development of its pipeline in 2027. The post Attovia’s Upsized IPO Brings In $289M to Fund New Kind of Biologic for Immune Disorders appeared first on MedCity News .
Sometimes, dealing with cybersecurity can feel like the boardwalk arcade game Whac-A-Mole. Just when you’ve addressed a potential threat with hardware, software and operating system patches, another threat comes from a direction you never anticipated. A new threat to many organizations’ cybersecurity is looming, and it could be the biggest one of all: quantum computing. The good news is that universities and other organizations can take steps now to protect their systems that might be vulnerable in a few years. Quantum computing, where processors use the principles of quantum physics, is…
Despite the silk masks, oils, and sleepy girl mocktails swirling around social media, one of the greatest causes of non-restorative sleep is rarely mentioned by wellness influencers online: actual sleep disorders. The post The Sleep Crisis We May Be Missing: Why Millions of Americans Remain Undiagnosed appeared first on MedCity News .
Healthcare leaders who shape that future need more than innovation. They need a visible, credible voice that helps the people around them — patients, clinicians, employees, partners, and investors — understand what the change is, why it matters, and why this is the right person to lead it. The post The Healthcare Leaders Driving Change Need to Get Better at Explaining It appeared first on MedCity News .
The technology that turns the most heads in K–12 schools tends to be the flashy equipment — interactive displays, 3D printers, powerful gaming computers, lifelike robots. But IT leaders know that the technology that matters most is often invisible. It’s a fast network designed to handle a classroom full of devices, clear policies that guide smart decisions, and cybersecurity strategies that guard student and institutional data. Click the banner below to find the tools you need to protect your district from cyberthreats.
Dabble is a writing tool that helps users organize, save, and produce long writing drafts, and is perfect for thesis projects, novels, and more.
Every day I come to work at Gettysburg College wearing a college-branded cap that I bought with a $25 gift card, presented to me by the Student Senate for leading an effort to improve student access to the campus Wi-Fi. The post How we recovered student trust after big Wi-Fi challenges appeared first on eCampus News .
QuantHealth raised $45 million in Series B funding to expand its AI platform, which simulates clinical trials before they run and predicts patient outcomes for pharma companies. The startup is aiming to cut the industry’s 90% trial failure rate and speed up the timelines for effective drugs to reach the market. The post QuantHealth Snags $45M to Simulate Clinical Trials Before Patients Ever Enroll appeared first on MedCity News .
arXiv:2606.08044v2 Announce Type: replace-cross Abstract: Safety evaluation of large language models (LLMs) is largely behavioral: a model is certified safe when it refuses harmful requests and answers benign ones. But refusing on the prompts an auditor happens to try does not show that the model is far from harmful behavior. Behavioral tests observe outputs; they do not measure how easily an intervention on the model turns a refusal into compliance. We call the gap between what static audits certify and what an intervention can reach the audit gap, and we show it is realizable: one can build a model that matches its safety-aligned base on every static audit yet gives way to a small, known perturbation of its internal state. We construct such dissociated models from three safety-aligned bases (Gemma 2 2B, Llama 3.2 3B, Qwen 2.5 3B) and audit the base, dissociated, and openly harmful models with the same soft interventions in parameter and latent space; the latent attacks are summarized
arXiv:2606.05799v2 Announce Type: replace-cross Abstract: Existing calibration methods for Large Language Models (LLMs) often overlook a critical dimension of trustworthiness: a model's behavioral robustness to irrelevant or misleading information. In this paper, we argue that a model's true confidence should reflect its stability under cognitive pressure. We introduce CaliDist, a novel post-hoc calibration approach that directly measures and penalizes a model's susceptibility to distraction. CaliDist quantifies how an LLM's predictions and uncertainty change when its input prompt is perturbed with semantic distractors. This stability (or lack thereof) signal is then used to adaptively scale the model's initial confidence score. Our extensive experiments on seven Natural Language Understanding classification benchmarks using six distinct LLMs show that CaliDist consistently achieves lower Expected Calibration Error (ECE) and Brier Score compared with strong baselines. Remarkably, our m
arXiv:2605.20244v2 Announce Type: replace-cross Abstract: We present Lean Refactor, a plug-and-play retrieval-augmented agentic framework for multi-objective, controllable, and version-robust refactoring of Lean proofs. LLM-generated proofs are notoriously correct-but-verbose and brittle across library versions, yet existing refactoring works overlook three practical challenges: 1) Lean refactoring is natively multi-objective (proof length, compilation cost, and version compatibility are often in tension); 2) Lean repositories have fragile compatibility, whereas LLM releases are unaware of Lean/Mathlib versions; 3) Training-based pipelines require repeated fine-tuning with each new LLM release, scaling neither with model churn nor with Lean's release cycle. Lean Refactor steers a frozen agentic LLM with retrievals from a curated database of multi-objective refactoring strategies, each densely annotated with metadata such as supported Lean/Mathlib versions and expected compilation-cost
arXiv:2604.23586v2 Announce Type: replace-cross Abstract: Joint audio-video generation models have shown that unified generation yields stronger cross-modal coherence than cascaded approaches. However, existing models couple modalities throughout denoising via pervasive attention, treating high-level semantics and low-level details in a fully entangled manner. This is suboptimal for talking head synthesis: while audio and facial motion are semantically correlated, their low-level realizations (acoustic signals and visual textures) follow distinct rendering processes. Enforcing joint modeling across all levels causes unnecessary entanglement and reduces efficiency. We propose Talker-T2AV, an autoregressive diffusion framework where high-level cross-modal modeling occurs in a shared backbone, while low-level refinement uses modality-specific decoders. A shared autoregressive language model jointly reasons over audio and video in a unified patch-level token space. Two lightweight diffusio
arXiv:2604.22207v2 Announce Type: replace-cross Abstract: Due to the textual and repetitive nature of many Requirements Engineering (RE) artefacts, Large Language Models (LLMs) have proven useful to automate their generation and processing. In this paper, we discuss a possible approach for automating the Goal-Oriented Requirements Engineering (GORE) process by extracting functional goals from software documentation through three phases: actor identification, high and low-level goal extraction. To implement these functionalities, we propose a chain of LLMs fed with engineered prompts. We experimented with different variants of in-context learning and measured the similarities between input data and in-context examples to better investigate their impact. Another key element is the generation-critic mechanism, implemented as a feedback loop involving two LLMs. Although the pipeline achieved 61% accuracy in low-level goal identification, the final stage, these results indicate the approach
arXiv:2604.11344v2 Announce Type: replace-cross Abstract: Embedding-as-a-Service (EaaS) has become an important semantic infrastructure for natural language and multimedia applications, but it is highly vulnerable to model stealing and copyright infringement. Existing EaaS watermarking methods face a fundamental robustness--utility--verifiability tension: trigger-based methods are fragile to paraphrasing, transformation-based methods are sensitive to dimensional perturbation, and region-based methods may incur false positives due to coincidental geometric affinity. To address this problem, we propose GeoMark, a geometry-aware localized watermarking framework for EaaS copyright protection. GeoMark uses a natural in-manifold embedding as a shared watermark target, constructs geometry-separated anchors with explicit target--anchor margins, and activates watermark injection only within adaptive local neighborhoods. This design decouples where watermarking is triggered from what ownership i
arXiv:2604.02486v3 Announce Type: replace-cross Abstract: Vision-language models (VLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they often fail on tasks that require fine-grained visual perception, even when the required information is still present in their internal representations. Prior work has attributed this ``hidden-in-plain-sight'' gap to the language model, but the cause remains unexplained. In this work, we demonstrate that this gap arises from the language model's lack of semantic labels for fine-grained visual details: when visual entities can be mapped to known concepts, VLMs bypass visual comparison and reason through language; when they cannot, VLMs resort to brittle and hallucinated descriptions. We verify this across semantic correspondence, synthetic shape matching, and face matching, and find that VLMs perform much better when the relevant entities are nameable than when they are unnamable. Mechanistically, Logit Lens an
arXiv:2603.24929v2 Announce Type: replace-cross Abstract: Understanding and quantifying uncertainty in large language model (LLM) outputs is critical for reliable deployment. However, traditional evaluation approaches provide limited insight into model confidence at individual token positions during generation. To address this issue, we introduce LogitScope, a lightweight framework for analyzing LLM uncertainty through token-level information metrics computed from probability distributions. By measuring metrics such as entropy and varentropy at each generation step, LogitScope reveals patterns in model confidence, identifies potential hallucinations, and exposes decision points where models exhibit high uncertainty, all without requiring labeled data or semantic interpretation. We demonstrate LogitScope's utility across diverse applications including uncertainty quantification, model behavior analysis, and production monitoring. The framework is model-agnostic, computationally efficien
arXiv:2603.19087v2 Announce Type: replace-cross Abstract: Creativity is the ability to come up with novel ideas, a capacity crucial for human development and flourishing. Are large language models (LLMs) creative in the same way humans are, and can the same interventions increase creativity in both? We study a promising but largely untested intervention for creativity: forcing creators to draw an analogy from a random, remote source domain (''cross-domain mapping''). Human participants and LLMs generated novel designs for ten daily products (e.g., backpack, TV) under two prompts: (i) cross-domain mapping, which required drawing inspiration from a randomly assigned source (e.g., octopus, cactus, GPS), and (ii) user need, which required proposing innovations targeting unmet user needs. We show that humans reliably benefit from randomly assigned cross-domain mappings, while LLMs, on average, generate more original ideas than humans and do not show a statistically significant effect of cro
arXiv:2511.03942v2 Announce Type: replace-cross Abstract: We present MIDI-LLM, a recipe that improves multitrack text-to-MIDI generation via adapting Large Language Models (LLMs). MIDI-LLM expands an LLM's text vocabulary to include MIDI tokens and employs a two-stage training pipeline: (i) unimodal continued pretraining on music-adjacent text and standalone MIDIs, and (ii) multimodal supervised finetuning on text-MIDI pairs. Our instantiation of MIDI-LLM based on Llama 3.2 (1B) outperforms the recent Text2midi model in both text control and musical quality, and readily integrates with optimized inference ecosystems like vLLM. To align with real-world songwriting workflows, we further finetune our MIDI-LLM on the TheoryTab dataset for text-conditioned lead sheet (i.e., melody + chords) generation and infilling. A comprehensive ablation study validates the synergy between LLM text pretraining, standalone MIDI pretraining, and supervised text-to-MIDI finetuning. Finally, an in-the-wild b
arXiv:2510.11834v3 Announce Type: replace-cross Abstract: Generative models are increasingly paired with safety classifiers that filter harmful or undesirable outputs. A common strategy is to fine-tune the generator to reduce the probability of being filtered, but this can be suboptimal: it often pushes the model toward producing samples near the classifier's decision boundary, increasing both false positives and false negatives. We propose Boundary Guidance, a reinforcement learning fine-tuning method that explicitly steers generation away from the classifier's margin. On a benchmark of jailbreak, ambiguous, and longcontext prompts, Boundary Guidance improves both the safety and the utility of outputs, as judged by LLM-as-a-Judge evaluations. Comprehensive ablations across model scales and reward designs demonstrate the robustness of our approach.
arXiv:2509.23690v2 Announce Type: replace-cross Abstract: Safety hazards in the home are a leading cause of preventable domestic injuries, motivating an automated inspector that actively explores a home and reports hazards before they cause harm. We introduce HomeSafeBench, the first benchmark for free-exploration home safety inspection with egocentric visual feedback, in which an embodied agent navigates a fully interactive 3D home, adjusts its viewpoint, and reports hazards purely from rendered first-person views. Built on the VirtualHome simulator, it covers five categories of common household hazards and comprises 1,000 human-validated inspection tasks. Evaluating a broad range of state-of-the-art Vision-Language Models (VLMs) reveals a large gap, where the best model reaches only about 34.7% F1, far below the 98.0% of a human inspector. Moreover, precision far exceeds recall across models, revealing a systematic tendency to under-report hazards that reflects a shared deficiency in
arXiv:2509.21012v4 Announce Type: replace-cross Abstract: In-context Learning (ICL) is an emerging few-shot learning paradigm based on modern Language Models (LMs), yet its inner mechanism remains unclear. In this paper, we investigate the mechanism through a novel perspective of information removal. Specifically, we demonstrate that in the zero-shot scenario, LMs encode queries into non-selective representations in hidden states containing information for all possible tasks, leading to arbitrary outputs without focusing on the intended task, resulting in near-zero accuracy. Meanwhile, we find that selectively removing specific information from hidden states by a low-rank filter effectively steers LMs toward the intended task. Building on these findings, by measuring the hidden states on carefully designed metrics, we observe that few-shot ICL effectively simulates such task-oriented information removal processes, selectively removing the redundant information from entangled non-select
arXiv:2508.20697v4 Announce Type: replace-cross Abstract: As large language models (LLMs) continue to grow in capability, so do the risks of harmful misuse through fine-tuning. While most prior studies assume that attackers rely on supervised fine-tuning (SFT) for such misuse, we systematically demonstrate that reinforcement learning (RL) enables adversaries to more effectively break safety alignment and facilitate more advanced harmful task assistance, under matched computational budgets. To counter this emerging threat, we propose TokenBuncher, the first effective defense specifically targeting RL-based harmful fine-tuning. TokenBuncher suppresses the foundation on which RL relies: model response entropy. By constraining entropy, RL-based fine-tuning can no longer exploit distinct reward signals to drive the model toward harmful behaviors. We realize this defense through entropy-as-reward RL and a Token Noiser mechanism designed to prevent the escalation of harmful capabilities. Exte
arXiv:2508.05149v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated potential in handling spoken inputs for high-resource languages, reaching state-of-the-art performance in various tasks. However, their applicability is still less explored in low-resource settings. This work investigates the use of Speech LLMs for low-resource Automatic Speech Recognition using the SLAM-ASR framework, where a trainable lightweight projector connects a speech encoder and a LLM. Firstly, we assess training data volume requirements to match Whisper-only performance, re-emphasizing the challenges of limited data. Secondly, we show that leveraging mono- or multilingual projectors pretrained on high-resource languages reduces the impact of data scarcity, especially with small training sets. Using multilingual LLMs (EuroLLM, Salamandra) with whisper-large-v3-turbo, we evaluate performance on several public benchmarks, providing insights for future research on optimizing S
arXiv:2507.19308v2 Announce Type: replace-cross Abstract: In this paper, we present our studies and experiments carried out for the task 1 of the Challenge and Workshop on Multilingual Conversational Speech Language Model (MLC-SLM), which focuses on advancing multilingual conversational speech recognition through the development of speech language models architectures. Given the increasing relevance of real-world conversational data for building robust Spoken Dialogue Systems, we explore three approaches to multilingual ASR. First, we conduct an evaluation of the official baseline to better understand its strengths and limitations, by training two projectors (linear and qformer) with different foundation models. Second we leverage the SLAM-ASR framework to train a custom multilingual linear projector. Finally we investigate the role of contrastive learning and the extended conversational context in enhancing the robustness of recognition.
arXiv:2605.30580v2 Announce Type: replace Abstract: Speculative decoding is a popular technique for large language model (LLM) inference, enabling faster generation by drafting multiple tokens with a smaller draft model. However, the effectiveness of speculative decoding has mainly been studied for English. Motivated by the curse of multilinguality, we hypothesize that speculative decoding is far less effective for low-resource languages due to the limited multilingual capacities of smaller models. We test eleven languages under a standard speculative decoding setup and find strong evidence for our hypothesis. Next, we try to improve the multilingual capabilities of the smaller draft model via distillation from the larger model. We find, though, that distillation generalizes poorly across tasks in the same language, and we argue that assembling a task-agnostic, fully representative dataset is infeasible for low-resource languages. Finally, we propose weaker n-gram models as draft model
arXiv:2605.28818v2 Announce Type: replace Abstract: Large language models have become increasingly useful computational models of human language processing, but it remains open whether vision-language learning makes text representations more human-like during natural reading. We address this question by comparing matched LLM and vision-language model pairs under strictly text-only input and evaluating alignment with human brain activity (whole-cortex fMRI) and human behavior (synchronized regressive saccades). We identify a selective, rather than global, effect of vision-language training on human-model alignment. In the two within-lineage model pairs, VLMs more accurately predicted human regressive saccades, whereas VLMs and LLMs showed comparable whole-cortex fMRI alignment. However, sentence-level analyses revealed that the VLM advantage in fMRI alignment increased with the visual evocative strength of the sentences. Together, these findings provide a controlled in-silico comparison
arXiv:2605.27642v2 Announce Type: replace Abstract: Soft prompting, also known as continuous prompting, is a parameter-efficient method for tuning LLMs to specific tasks. Like other machine learning techniques, its parameters encode some hidden procedure: is it possible to train a model to decode this procedure---to "translate" raw parameters into natural language? In this work, we present a promising proof-of-concept: a translator model capable of verbalizing soft prompt's learned embeddings into fluent natural language descriptions. We show that these verbalizations when used as standalone prompts for inference surpasses baselines, suggesting that they are not just plausible-sounding descriptions, but genuinely relevant to the task. On average, verbalizations retain a modest but significant 32\% of the original soft prompt's performance. We speculate on future directions for how this could be used for interpretability or inference or perhaps even extended to other ML techniques.
arXiv:2605.26275v2 Announce Type: replace Abstract: Automatic prompt engineering (APE) rewrites prompts to improve downstream task performance, but existing APE loops treat the optimizer itself as a fixed pipeline. We port the code-as-action paradigm of CodeAct (Wang et al., 2024a) to APE and propose SPEAR (Sandboxed Prompt Engineer with Active Roll-back), a free-form agentic optimizer with four tools -- evaluate, python, set_prompt, finish -- that decides autonomously how and when to use them. The distinctive tool is the Python sandbox: the optimizer writes and executes arbitrary Python on the current evaluation DataFrame, performing structural error analysis (confusion matrices, error clustering, per group metrics) the agent itself authors. Two guardrails turn the long-horizon agent into a monotone-improving optimizer: auto-rollback on metric regression, and an optional guard metric floor. We evaluate on three industrial LLM-as-judge suites (13 judge tasks across recruiter-intake, co
arXiv:2604.13197v3 Announce Type: replace Abstract: Process reward models (PRMs) provide fine-grained supervision for reasoning, but reliable PRMs often require step annotations or heavy verification pipelines, making them costly to scale and refresh during online RL. Implicit PRMs reduce this cost by training log-likelihood-ratio rewards from trajectory-level outcome labels. However, the log-ratio is constrained only as a sequence-level aggregate during training, while inference decomposes it into token- or step-level scores for partial prefixes. This train-inference mismatch leaves local credits weakly identified, so distribution-wide scoring can amplify misleading advantages. We propose Implicit Prefix-Value Reward Model (IPVRM), which directly learns the probability of eventual correctness for each prefix from outcome labels. Step signals are then obtained as temporal-difference (TD) differences between consecutive prefix values, aligning the training target with inference-time use
arXiv:2604.06374v2 Announce Type: replace Abstract: Latent reasoning via continuous chain-of-thoughts (Latent CoT) has emerged as a promising alternative to discrete CoT reasoning. Operating in continuous space increases expressivity and has been hypothesized to enable superposition: the ability to maintain multiple candidate solutions simultaneously within a single representation. Despite theoretical arguments, it remains unclear whether language models actually leverage superposition when reasoning using latent CoTs. We investigate this question across three regimes: a training-free regime that constructs latent thoughts as convex combinations of token embeddings, a fine-tuned regime where a base model is adapted to produce latent thoughts, and a from-scratch regime where a model is trained entirely with latent thoughts to solve a given task. Using Logit Lens and entity-level probing to analyze internal representations, we find that only models trained from scratch exhibit signs of u
arXiv:2604.06005v2 Announce Type: replace Abstract: Interpreting the information encoded in language model weights remains a fundamental challenge in mechanistic interpretability. In this work, we introduce ROTATE (Rotation-Optimized Token Alignment in weighT spacE), a data-free method requiring no forward passes that disentangles MLP neurons directly in weight space. Our approach relies on a key statistical observation: neurons that encode coherent, monosemantic concepts exhibit high kurtosis when projected onto the model's vocabulary. By optimizing rotations of neuron weights to maximize their vocabulary-space kurtosis, our method recovers sparse, interpretable directions which we name vocabulary channels. Experiments on Llama-3.1-8B-Instruct and Gemma-2-2B-it demonstrate that ROTATE consistently recovers vocabulary channels that are faithful to the neuron's behavior; ablating individual channels selectively disables corresponding input activations or the promotion of specific concep
arXiv:2604.04790v2 Announce Type: replace Abstract: Natural language processing (NLP) advances have powered a generation of LegalTech systems, but Turkish law remains under-served by domain-specific data and models. English has legal encoders such as LEGAL-BERT; no comparable high-volume Turkish counterpart exists. We introduce HukukBERT, a Turkish legal language model trained on a 19 GB cleaned corpus using a hybrid domain-adaptive pre-training (DAPT) recipe that mixes Whole-Word Masking, Token Span Masking, Word Span Masking, and targeted Keyword Masking. We compared our 48K WordPiece tokenizer and DAPT pipeline against general-purpose and existing domain-specific Turkish models. On the Legal Cloze Test - a masked legal term prediction benchmark over Turkish court decisions - HukukBERT reaches 84.40% Top-1 accuracy and beats every baseline we tested. The Legal Cloze Test is synthetically constructed, so its passages are absent from the pre-training corpus by construction, eliminating
arXiv:2603.24413v2 Announce Type: replace Abstract: Poetry generation in Sanskrit typically requires the verse to be semantically coherent and adhere to strict prosodic rules. In Sanskrit prosody, every line of a verse is typically a fixed length sequence of syllables adhering to prescribed binary patterns of syllable weights. We observe that instead of treating a verse as a monolithic sequence, segmenting them as grouped-lines leads to significant improvement in semantic coherence by 10\% with comparable metrical adherence. Specifically, Pingala, our proposed decoding approach is designed to encourage every line to have well-formed words and our token selection biases the model towards it by preferring longer tokens. Writing in Sanskrit follows phonemic orthography, hence using a phonetically aware transliteration scheme, SLP1, increased the metrical alignment by 46\% with comparable semantic similarity, for a instruction fine-tuned large language models like Phi-4. We also introduce
arXiv:2603.19066v2 Announce Type: replace Abstract: Four-term word analogies (A:B::C:D) are classically modeled geometrically as parallelograms: adding the vector B-A+C produces D. Recent work suggests that this model poorly captures how humans produce analogies, with simple local-similarity heuristics often providing a better account (Peterson et al., 2020). But does the parallelogram model fail because it is a bad model of analogical relations, or because people are not very good at generating relation-preserving analogies? We compared human and large language model (LLM) analogy completions on the set of problems from Peterson et al. (2020). We find that LLM-generated analogies are reliably judged as better than human-generated ones, and are also more consistent with parallelograms in a distributional embedding space. Crucially, we show that the improvement over human analogies is driven by greater parallelogram alignment and reduced reliance on accessible words rather than enhanced
arXiv:2603.10771v2 Announce Type: replace Abstract: Large language models (LLMs) trained with canonical tokenization exhibit surprising robustness to non-canonical inputs such as character-level tokenization, yet the mechanisms underlying this robustness remain unclear. We study this phenomenon through mechanistic interpretability and identify a core process we term word recovery. We first introduce a decoding-based method to detect word recovery, showing that hidden states reconstruct canonical word-level token identities from character-level inputs. We then provide causal evidence by removing the corresponding subspace from hidden states, which consistently degrades downstream task performance. Finally, we conduct a fine-grained attention analysis and show that in-group attention among characters belonging to the same canonical token is critical for word recovery: masking such attention in early layers substantially reduces both recovery scores and task performance. Together, our fin
arXiv:2603.09986v3 Announce Type: replace Abstract: Hallucinations, the tendency for large language models to provide responses with factually incorrect and unsupported claims, is a serious problem within natural language processing for which we do not yet have an effective solution to mitigate against. Existing benchmarks for medical QA rarely evaluate this behavior against a fixed evidence source. We ask how often hallucinations occur on textbook-grounded QA and how responses to medical QA prompts vary across models. We conduct two experiments, the first experiment to determine the prevalence of hallucinations for a prominent open source large language model (LLaMA-70B-Instruct) in medical QA given closed-source zero-shot prompts, and the second experiment to determine the prevalence of hallucinations and clinician preference to model responses. We observed, in experiment one, with the passages provided, LLaMA-70B-Instruct hallucinated in 19.7\% of answers (95\% CI 18.6 to 20.7) even
arXiv:2603.01243v2 Announce Type: replace Abstract: Large language models (LLMs) are powerful tools that have found applications beyond human-machine interfaces and chatbots. Beside free-form generation, there has been an interest in constrained generation, a setting where LLMs are constrained to generate well-formed outputs with respect to the language defined by a formal grammar. Although appealing, this setting may be over restrictive for downstream applications. For example, many LLM tasks require the model to reason freely before generating its final response in a specific format. In this work, we introduce suffix-constrained generation, a constrained generation setting in which only the end of the response is constrained by a grammar, a scenario that is not supported by existing constrained generation methods. We introduce several suffix-constrained generation algorithms that are based on greedy search. We experiment on several datasets, and show that our approach allows to guara
arXiv:2601.16217v2 Announce Type: replace Abstract: Large language models increasingly mediate multilingual professional communication, where useful generation requires adapting to community conventions about which expressions are retained, translated, or mixed. Existing benchmarks rarely isolate such community-conditioned choices. We introduce ChiEngMixBench, a controlled benchmark for Chinese AI/CS discourse, where Chinese frames routinely incorporate established English technical terms. Built from public technical discussions, it contains 1,706 source-derived candidate pairs covering 1,344 non-empty normalized terms, including a 1,167-pair strict subset that fixes the Chinese prefix and syntactic position while varying only the terminology form. The benchmark combines paired likelihood comparisons with a transparent reference-profile diagnostic for open-ended responses. Across nine open-weight models, Chinese equivalents receive higher likelihood on most pairs, revealing a gap betwe
arXiv:2601.09445v2 Announce Type: replace Abstract: In language models (LMs), intra-memory knowledge conflict arises when inconsistent information about the same subject is encoded within the model's parametric knowledge. Prior work has primarily focused on resolving conflicts between a model's internal knowledge and external sources, which is known as context-memory knowledge conflict, through approaches such as fine-tuning or knowledge editing, while the understanding of conflicts that arise internally remains largely unexplored. In this work, we design a framework to identify where internal conflicting knowledge is encoded within LMs. We test our framework on four LMs using both synthetic and real-world knowledge conflicts. We find that internal conflicts often arise and are resolved in the final layers across all models, but that interventions are markedly less effective on real-world knowledge conflicts. Targeted attention-head interventions outperform layer-wise ones, and a filte
arXiv:2511.21214v4 Announce Type: replace Abstract: Explicit safety policies can improve reasoning-model safety, but their effective coverage may lag behind evolving jailbreak strategies. We study whether a reasoning model can synthesize and internalize a task-specific safety guideline from a small set of harmful and benign examples. We introduce Self-Guided Adaptive Safety Alignment (SGASA). The model generates a guideline, refines it on its own errors, and selects a version by self-evaluation, which can then be applied in context or distilled into the model for guideline-free inference. Across two adversarial prompt datasets and three Qwen3 scales, in-context guidelines improve a combined safety and non-over-refusal score by 20.0-45.5 points on the 8B and 14B models. Self-evaluation selects the externally best refinement round in five of six settings, while guideline generation and utilization show distinct scaling patterns. Using alignment supervision derived only from WildJailbreak
arXiv:2511.10696v3 Announce Type: replace Abstract: Sparse attention is crucial in long-context Transformers, which restricts each token to a limited neighborhood and thereby reduces the quadratic cost of full self-attention. Local windows capture nearby context effectively, yet they induce a receptive-field bottleneck for dependencies beyond the window, limiting long-range modeling under moderate depth. In this paper, we propose $\pi$-Attention, an \emph{online efficient} sparse attention operator: as tokens arrive, each step maintains a streaming working set of local neighbors plus a $\pi$-indexed long-range fetch, fused by an adaptive prior under a shared softmax. Rather than materializing a global sparse mask in advance, $\pi$-Attention computes attention on the live working set with hierarchy-aware IO. We analyze causal reachability and minimum depth under this online rule, and show per-step cost remains $\mathcal{O}(k)$. Experiments on language modeling, Long Range Arena, and eff
arXiv:2510.22014v2 Announce Type: replace Abstract: Discrete optimization-based jailbreaking attacks on large language models aim to generate short, nonsensical suffixes that, when appended onto input prompts, elicit disallowed content. Notably, these suffixes are often transferable -- succeeding on prompts and models for which they were never optimized. And yet, despite the fact that transferability is surprising and empirically well-established, the field lacks a rigorous analysis of when and why transfer occurs. To fill this gap, we identify three statistical properties that strongly correlate with transfer success across numerous experimental settings: (1) how much a prompt without a suffix activates a model's internal refusal direction, (2) how strongly a suffix induces a push away from this direction, and (3) how large these shifts are in directions orthogonal to refusal. On the other hand, we find that prompt semantic similarity only weakly correlates with transfer success. Thes
arXiv:2510.10961v4 Announce Type: replace Abstract: As language models become increasingly deployed in online environments, toxicity detection and detoxification have received growing attention. Existing studies primarily focus on non-obfuscated text, which limits robustness when users intentionally disguise toxic expressions. In particular, Korean toxic expressions can be easily disguised through agglutinative morphology and Hangeul-specific orthographic variation. However, obfuscation in Korean remains largely unexplored, which motivates us to introduce a KOTOX: Korean toxic dataset for deobfuscation and detoxification. We categorize Korean obfuscation patterns into linguistically grounded classes, define transformation rules derived from real-world examples, and provide the resulting obfuscation framework as an open transformation package. Using these rules, we provide paired neutral and toxic sentences alongside their obfuscated counterparts. Models trained on our dataset better ha
arXiv:2509.25459v2 Announce Type: replace Abstract: Large Language Models (LLMs) show promise in generating long-form scientific explanations that synthesize evidence and connect multiple factors. However, in long-form scientific question answering, LLMs often hallucinate, producing unsupported or inconsistent claims. Retrieval-Augmented Generation (RAG) improves trustworthiness by grounding generation in external sources; scientific simulators are valuable because they can validate quantitative hypotheses and capture evolving dynamics. Yet simulation-based RAG is non-trivial due to two challenges: how to retrieve from scientific simulators, and how to efficiently verify and update long-form answers. To overcome these challenges, we propose SimulRAG, a simulator-based RAG framework with a generalized retrieval interface that translates between text and simulator parameters/outputs. SimulRAG further introduces claim-level generation with uncertainty estimation and simulator boundary ass
arXiv:2508.12448v2 Announce Type: replace Abstract: In-context learning (ICL) lets large language models (LLMs) solve new tasks from prompts alone, across an ever-widening range of domains, yet the mechanisms underlying this ability remain poorly understood. Physical systems offer a controlled testbed for this question as they provide experimentally controllable data with structured dynamics grounded in fundamental principles. Here we study the ICL ability of LLMs, focusing on physical reasoning. Using dynamics forecasting as a proxy task, we first show that LLMs forecast physical dynamics in context, with accuracy improving as more history is provided. Analyzing the model's residual stream reveals internal activations that correlate with key physical quantities such as energy. These correlations strengthen gradually with context length, indicating that LLMs spontaneously form representations aligned with physical concepts without any physics-specific supervision. To assess whether the
arXiv:2506.13502v3 Announce Type: replace Abstract: Next-word prediction (NWP) trains language models against a single observed continuation, even though many contexts admit multiple plausible next words. Recent RL-based next-word reasoning methods make this tension explicit: they reward a model for producing a rationale that supports one context-conditioned continuation, which can turn a pre-existing preference into a confident, self-justifying trajectory. We introduce BOW, an RL framework that instead trains models to produce self-contained, neutral, and comprehensive descriptions of the plausible next-word space. BOW's core reward is mediated by the generated trajectory. The policy conditions on the full context, but a frozen scorer assigns the core reward from the trajectory alone, without receiving the original context as a separate input. The trajectory may restate relevant context; the bottleneck is the missing direct context-to-scorer path in the core reward. BOW-Reg adds a lig
arXiv:2407.11421v2 Announce Type: replace Abstract: Large Language Models (LLMs) exhibit emergent abilities that may reveal aspects of their internal mechanisms. We study one such capability: directly performing extended sequences of calculations without generating chain-of-thought solutions. The strongest models in our evaluation can directly output sums with up to 15 addends, where operands are sampled from 1 to 100. We hypothesize that models form Implicit Discrete State Representations (IDSRs) within their hidden states and use them for internal symbolic calculation. We test for these representations, characterize their formation from layer, digit, and sequence perspectives, and investigate their use in producing answers. We also find that these state representations are far from lossless in current open-source models, contributing to errors in final outputs. Our work offers an initial exploration of LLMs' symbolic calculation abilities and underlying mechanisms. Code and reproduci
arXiv:2608.04010v1 Announce Type: cross Abstract: Existing scaling strategies for Multimodal Large Language Models (MLLMs) typically expand either model parameters or sequential inference computation, incurring substantial memory or latency overhead. More importantly, most existing methods fail to alter the rigid, fixed computation allocation between the Vision Transformer and the Large Language Model components, limiting task-specific optimization. To address this, we introduce the Parallel Vision-Language (ParVL) scaling framework for MLLMs, which scales parallel computation by reusing the existing ViT and LLM backbone parameters across multiple vision and language branches. This framework raises a central question: given a fixed backbone parameter budget, how should additional shared-backbone computation be allocated between the vision and language modalities? We instantiate each parallel computational stream with branch-specific prefix parameters over a shared backbone, and train t
arXiv:2608.03999v1 Announce Type: cross Abstract: Text-to-music language models begin with a choice usually made by default: how to tokenize music. Normally entangled with backbone, data, and recipe, its effect has never been measured in isolation. We fix pretrained Qwen3.5 (0.8B-27B), data, budget, and decoding, and swap only the representation across seven tokenizations, anchoring texture metrics to each representation's model-free ceiling. The ordering is clean and surprising: representation, not model size, is the binding variable for distributional fidelity. Scaling the backbone 34x barely moves Frechet Music Distance (FMD), whereas switching representation halves it. PMT, a performance-resolution stream we release (10 ms timing, per-note velocity, multi-track texture; 609 symbols), reaches FMD 159 at 0.8B against 272-286 for beat grids (1.7-1.8x lower, up to 2.8x elsewhere; non-overlapping bootstrap CIs), so a 0.8B performance-resolution model beats a 27B beat grid. It reappears