People write and sketch while speaking to explain, organize, and develop content together. Inspired by these practices, we investigate how voice agents can use a canvas alongside speech in multi-turn conversations with users. We conducted a two-part formative study: an observational study of how pairs coordinated speech and boardwork, followed by a design workshop that informed a design space for speech-canvas interaction with voice agents. Building on these insights, we developed VoCa, a voice agent that coordinates speech with visual object creation, annotation, and attention guidance. A five-day deployment with 18 participants examined usability, experiences of speech-canvas interaction, patterns of use, and desired improvements. Participants' experiences highlighted opportunities for speech-canvas interaction in learning, work, and daily life, alongside challenges in coordinating what agents say and show in ways users can follow and influence. These findings inform how voice agents can use a canvas alongside speech in conversation.
Automated neurodevelopmental assessment increasingly combines computer vision, speech, eye tracking, physiology, and machine learning, yet multimodality and discrimination do not establish construct validity, clinical usefulness, or appropriate interpretation of behavioral variation. We propose a testable architecture for neurodiversity-aware multimodal affective computing in which population-relative deviation is not treated as sufficient evidence of adverse functioning. This conceptual article reports neither a new participant-level study nor a trained predictive system, and no predictive or clinical superiority is claimed. The framework keeps observable signals, modality quality and availability, context, population-relative information, pooled within-person references and, when sufficiently supported, context-conditioned within person references, and predictive uncertainty distinguishable throughout inference. It defines structured decision-support outputs and auditable requirements for target validity, temporal integrity, contextual leakage, missing-modality robustness, calibration, subgroup evaluation, interpretability, privacy, and human decision boundaries. The contribution is architectural rather than algorithmic: it specifies empirically testable constraints that can be instantiated with different multimodal learning methods. Autism provides the principal motivating evidence base, without assuming unchanged transfer to other neurodevelopmental conditions.
Assine o Pro, sem anúncios
Faça upgrade para uma leitura sem interrupções e acesso prioritário a novas fontes.
Recently proposed Drifting Models shift iterative distribution refinement from inference to training, enabling effective one-step generation. However, their performance on complex image datasets depends strongly on the representation used to construct the drifting field: pixel-space drifting performs poorly, whereas pretrained feature spaces substantially improve sample quality for reasons that remain unclear. We trace this gap to the discriminative geometry of the representation, which determines sample weighting in kernel density estimation (KDE) and, consequently drift. We introduce persistent representation learning, which continuously learns a more discriminative representation geometry as the generator evolves across batches. We further establish a current-step gradient equivalence between the KDE ratio loss and drift regression loss under matched conditions, connecting density-ratio-based generator optimization to empirical drifting and motivating direct control of the drifting velocity. Across multiple datasets, our method learns effective discriminative representations directly from pixels and reduces FID by approximately 82-95\% over the original pixel-space Drifting Models, without pretrained encoders. Adapting pretrained representations and applying velocity clipping provide further gains.
A verifier for an agent faces rules of two kinds: the ones a fixed check can settle and the ones that require a judge. A team that derives its own checks fixes that split up front. Where the requirements come from outside, as in finance, healthcare and law, the agent enforces rules it did not write, so the split falls to runtime, recurring for every predicate of every rule on every action at a rate no reviewer can audit. Every escalation scheme assumes a model can make that decision itself, that it is self-decidable. Across six corpora, including the EU AI Act, FINRA guidance and a deployed credit agent, we collect roughly 22,000 labels from four models built by three labs. They agree almost perfectly where the answer is obvious and collapse on regulatory text; their errors run in opposite directions, so no model can be trusted as the conservative choice; and on the deployed agent's own rule-set they err together, over-claiming that a fixed check will do, the direction that never gets escalated. We introduce CoVer (corroborate-then-verify), which treats unanimity as a nomination, admitting a predicate only when the check synthesized for it survives intervention, reading fields the agent cannot write and holding under deterministic rewording. That gate rejects most of what corroboration wrongly admits, at a cost in coverage we report rather than tune away. The obvious alternative, agreement with a reference judge, certifies nothing: it climbs from 30% to 77% across calibration bands while the genuinely decidable share does not move, because a judge drawn from the population under indictment ratifies the blind spot it shares. Self-decidability is not a capability to elicit from a model but a boundary the verifier must construct.
Diffusion and flow models provide expressive policy classes for online reinforcement learning (RL), enabling multimodal behaviors and improved performance. However, training these policies remains challenging: the critic specifies the desired policy as an unnormalized Boltzmann density but does not provide direct samples from it. Many existing methods rely on importance sampling to construct training signals, which can suffer from high variance, increasing computational cost and destabilizing training. We propose Score-Calibrated Flow (SCF), a simple and efficient algorithm for training generative models to sample from unnormalized densities without importance sampling or backpropagation through the sampling trajectory. We learn the desired flow by enforcing self-consistency, bypassing target posterior mean estimation. By jointly exploiting the prescribed target score and the structure of flow matching, we establish these self-consistency requirements as score-calibrated optimality conditions, first for the terminal density and then for the trainable velocity field. We prove that their unique solutions are, respectively, the target density and the ideal flow model that conditional flow matching (CFM) would recover if target samples were available. We formulate the velocity condition as a fixed-point equation and exploit its conditional-expectation structure to construct a stop-gradient objective for enforcing it. The resulting training procedure retains the scalable sample-interpolate-regress structure of CFM despite the absence of target samples, using endpoints generated by the current flow. For online RL, the critic gradient supplies the target score at the generated actions, yielding a direct approach to actor training. Experiments on RL benchmarks demonstrate that SCF matches or improves upon state-of-the-art generative-policy baselines, while substantially reducing training time.
Operator eigenvalue problems appear throughout science. Classical methods usually discretize the operator into a matrix and then solve the resulting matrix eigenvalue problem. This works well in low dimensions, but fine grids quickly become expensive in both memory and computation as the dimension grows. Neural network based solvers avoid storing these large grids, but recent state of the art neural methods can require hundreds of thousands of training steps and may struggle to find the desired eigenvalues. We show that the two approaches can help each other. A coarse finite difference method (FDM) calculation acts as a cheap numerical model of the operator spectrum. We use the approximate eigenvalues as fixed shifts during the training of the neural solver, as they only need to locate the relevant part of the spectrum. We also introduce Stabilized Inverse Power Method Neural Network (SIPMNN), a more stable training procedure for higher-dimensional problems. Across five test problems at d=10, the combined approach is more accurate overall than the tested fully neural alternatives while using eight to ten times fewer iterations.
Agents are entering finance, healthcare and law, sectors where a violation leaves no lexical signature and carries real penalties. Whether an omission is material, or a disclosure sufficient, depends on what the response left out. Probing such an obligation means finding responses one minimal edit from flipping compliance, and every probe costs a generation and two adjudications, so the binding constraint on regulatory red-teaming is sample efficiency, not volume. We introduce Penumbra, an adversarial search that walks from a verified anchor under an expanding edit budget until a two-evaluator committee changes its verdict, and emits the two adjacent responses that straddle the change. Allocation is adaptive, and the objective is coverage of the defeat surface: distinct (obligation x defeat mode) cells resolved per candidate. At matched budget, adaptive allocation reaches uniform allocation's full-budget coverage on 59% of the candidates; at equal records it covers 1.43x the defeat modes of naive enumeration, and the gain is confined to the axis it targets. On 60 screened obligations of a financial advisory constitution, Penumbra returns 144 pairs, each a compliant and a violating response that the committee places on opposite sides of the boundary, differing by a handful of words where a model asked for both directly produces texts sharing almost nothing. A second constitution, for clinical triage, reproduces this on 18 obligations: 49 pairs at the same tightness, with the same modes hardest. Pairs like these show where an obligation's own terms stop deciding, which is what an agent deployed under it must be tested against, and the search finds them at a cost that scales with the boundary, not the text.
Retrieval-Augmented Generation (RAG) is typically evaluated under the implicit assumption that the underlying knowledge base (KB) is clean, leaving the behaviour of RAG systems under realistic KB degradation poorly characterised. We introduce RAGStress, a controlled evaluation benchmark for stress-testing RAG systems under systematic KB corruption. The benchmark pairs four naturalistic corruption types (factual corruption, numeric typo, relevance poisoning, and contradiction injection) with three severity levels (subtle, moderate, and obvious) over a single-KB, metadata-filtered experimental design built from 57 MMLU subjects and 182,546 documents. Across 52,500 model-question-condition evaluations, RAGStress reveals that clean retrieval can mask robustness differences, semantic-fidelity corruptions are substantially more harmful than signal-utility perturbations, no-retrieval accuracy does not predict corrupted-retrieval robustness, and mixed-KB accuracy should not be treated as worst-case robustness. We document the benchmark's intended use, supported claims, and limitations, and provide an artifact bundle including generation scripts, corruption prompts, metadata schema, and evaluation code. RAGStress is intended as a controlled stress test for RAG robustness under KB corruption, not as a general model leaderboard.
Self-supervised learning (SSL) is standard for speech representation learning, but mainstream models are designed around single-speaker audio, limiting their usefulness in multi-speakers scenarios. We present SepRQ, an open-source SSL framework that replaces masked prediction with a pseudo-source-separation objective over frozen random-projection codebooks. By adopting a novel mask-free, multiresolution approach, SepRQ achieves state-of-the-art performance in Speaker Diarization and Speech Separation on the SUPERB benchmark, surpassing WavLM and other cocktail-party derived SSLs at both Base and Large scales, while requiring only 85.68M inference parameters. SepRQ also demonstrates strong performance across target-speaker tasks requiring enrollment (such as Target-Speaker Automatic Speech Recognition), and on the challenging multi-domain DIHARD 3 diarization dataset. Notably, we report strong separation capabilities on three-speaker mixtures (WSJ0-3Mix), where current SSL literature struggles. While cocktail-party SSLs remain scarce and closed-source, limited to C-HuBERT and the enrollment-based SA-WavLM, we open-source SepRQ to the community.
We investigate error discovery and handling in question answering over imperfect tables through controlled studies across three large language models (LLMs) on human-reviewed RADAR-T examples. Answering questions over these tables requires handling errors that can affect the answer. We vary row order and compare original, error-marked, and repaired tables to test whether discovery depends on where errors appear and whether providing their locations is sufficient for accurate question answering. First, reordering rows changes error discovery even when the table contents and gold answer remain unchanged. Complete discovery is higher for back than front placements and, averaged over the tested mean positions, for compact than widely spaced layouts. Second, providing verified error locations alone is insufficient for accurate QA, leaving a substantial accuracy gap between error-marked and repaired tables. Providing tables with human-reviewed repairs already applied raises code-assisted QA accuracy by 39.0-59.1 percentage points over the error-marked tables across the three systems. GBDI, a simple workflow, puts these findings into practice by combining error discovery across shuffled table views with explicit guidance for verifying and handling the reported errors. On RADAR-T, GBDI raises observed QA accuracy by 3.8-18.5 percentage points over a code-agent baseline across five systems. These results highlight the importance of both reliable error discovery and effective error handling in question answering over imperfect tables. Our anonymous repository is available at https://anonymous.4open.science/r/GBDI-ICLR-2027-85BD/
Multi-agent debate can improve reasoning, yet often fails to beat simple majority voting. We argue that successful debate requires two distinct mechanisms: proposal supply must surface a correct answer, and readout must identify that answer when voting misses it. We formalize the first requirement through recoverable headroom, which measures cases where a correct proposal is available but the majority answer is wrong. For the second, we develop Latent Verification Debate (LVD), an accounting model in which candidate proposals receive answer-specific verification evidence before final generation. Controlled fixed-proposal interventions estimate this latent effect in equivalent peer-support units and show that correct evidence changes answer probabilities and generated decisions while proposal supply remains fixed. To improve proposal supply, we construct societies from neural-thicket agents using labeled and label-free coverage objectives. Across two backbones and matched-budget reasoning benchmarks, coverage-selected societies increase complementary proposal supply and improve aggregate accuracy in repeated stochastic evaluations. Round-level controls further show that interaction provides gains beyond applying the same finalizer directly to the initial proposals. These results identify proposal coverage and truth-sensitive evidence use as complementary conditions for debate to outperform voting. Code is available at https://github.com/Wang-ML-Lab/when-debate-helps.
Activation steering has proven effective for controlling the behavior of Large Language Models (LLMs) at inference time, but its application to SpeechLLMs remains new, and training-free steering approaches for such models are still largely unexplored. We propose a training-free Contrastive Activation Addition (CAA) protocol that derives steering vectors for common speech tasks (e.g. transcription) in SpeechLLMs from a small number of labeled utterances. We showcase that adding these vectors in the representation space, at inference time, enforces better the targeted speech task. We further show that, when combined with prompting, these vectors yield to consistent improvement over prompting alone on most evaluated tasks such as Automatic Speech Recognition (ASR) or Emotion Recognition (ER), and transfer to out-of-domain data. We additionally demonstrate the usefulness of script-normalization directions to enforce the target script of a specific language.
Large Language Models (LLMs) can adopt distinct personas to tune their semantics, expertise, and perspective to different users and tasks. Precise control over these traits is critical to ensure safety and reliability in model behavior. Existing methods like activation steering and prompt-based persona induction reduce a persona to a single dominant direction, missing the finer, nested traits that emerge only once that dominant signal is factored out. We introduce modulation as a setting where the persona context is already embedded in the content being manipulated, requiring control methods to amplify or suppress a trait already present rather than inject it from scratch. PaSS is an inference-time control paradigm that models personas as multi-dimensional subspaces in a model's latent space without supervised contrastive examples. The persona subspaces are extracted via iterative concept erasure and applied to modulate persona-guided generation without retraining. To extract this subspace, we use Iterative Nullspace Projections (INLP) to linearly and iteratively isolate persona-specific directions. We causally evaluate six personas against diverse tasks like MATH-500, TinyAlpaca, GSM8K, and IFEval, showing that discriminative, iterative subspace extraction captures diverse traits underlying a given persona, enabling stronger and larger modulation than single-direction additive methods, while maintaining content fidelity. We further study individual peeled directions within each subspace to uncover the distinct aspects of persona behavior they encode. Overall, we show that persona subspaces offer a controllable, interpretable, and generalizable framework for modulating LLM behavior without sacrificing task performance.
Large language models (LLMs) have progressively evolved into the core of autonomous agents. Building on this progress, LLM-based multi-agent systems (MAS) coordinate multiple agents into a synergistic team to accomplish complex tasks that exceed the capabilities of individual agents. The effectiveness of such systems depends not only on the agents themselves, but also on how collaboration mechanisms are designed and organized. Note that real-world collaboration is typically partially observable, where each agent can only access partial information about the environment due to physical or privacy-related constraints. However, many existing multi-agent benchmarks assume global observability, and leave limited support for systematically evaluating collaboration mechanisms. To bridge this gap, we introduce MASBench, a multi-agent collaboration benchmark designed under partially observable constraints. It is organized into three progressive task categories: Reasoning, Scheduling, and Game. Through this structure, we progressively evaluate three representative collaboration mechanisms: Protocol, Memory, and Routing. MASBench further provides deterministic evaluation metrics, including performance score, communication cost, and cost effectiveness, to characterize both collaboration outcomes and communication overhead. Experiments across diverse LLM backbones and mechanism configurations offer empirical guidance for effective MAS design. Code is available at: https://github.com/BUPT-GAMMA/MASBench
A federated objective is a weighted sum of client risks, and the weights are almost always fixed in advance. We treat them instead as the only instrument of a wisdom-of-crowds mechanism: clients are noisy views of one truth, each seeing it through an independent distortion that is unbiased across the crowd. That the optimal weights are inversely proportional to the clients' error energies is classical; we begin at the question that answer presupposes, which energies belong there and whether a crowd can recover them from itself. Excess risk on the truth is of the exact order of the aggregate bias energy, so no optimizer can repair a bad weight vector; the truth itself is identifiable only up to a linear tilt, so subtracting estimated client biases provably reproduces uniform weighting. The expected per-client second moments, however, are exactly identified from the law of the crowd's disagreement by a well-conditioned linear inversion, a step random-effects meta-analysis cannot take because a source reports once; their realized counterparts are estimable up to an incoherence floor the algorithm can measure. This yields CROWD, which reads the disagreement off the optimization trajectory at no extra cost and matches a Bayesian minimax lower bound in the same constant: per instance as the horizon grows, and unconditionally as the prior becomes diffuse. For arbitrary distortions it stays competitive with the optimal weights, at a ratio governed by a geometric incoherence the algorithm can measure. On real scans split into sites with their own miscalibrated detectors it attains the oracle excess risk; on a companion federation that pulls bias and noise apart, weighting by noise variance is worse than not weighting at all, and CROWD is not.
Fine-tuning a language model on new text degrades what it already does. Replay-free projectors such as Adam-NSCL and GPM forbid one shared subspace of a layer's inputs in every row of the update. The tropical geometry of a ReLU layer shows why this is too coarse. In data space, the units' walls are tropical hypersurfaces whose cells are dual to the upper vertices of a zonotope; in weight space, each old token is a hyperplane, and the tokens cut out a polyhedron, the closure of the weights that keep every token on its side. An exact identity joins the two pictures: the squared change of the layer's output under any weight change splits into in-cell, open-to-closed and closed-to-open terms, and the first two live on the tokens each unit fires on. The identity names a gate-aware per-unit projector, and a budget-separation theorem prices exact protection: it costs a unit the rank of its own open tokens, while a shared subspace pays at least the rank of their union in every row. On OPT-1.3b, where 96% of (token, unit) pairs are closed, the projector forgets less than Adam-NSCL at all six matched budgets from 9 to 60 constrained directions per row, the gap widening from 1.1to 4.3; with 1/5.5 of the directions it halves the forgetting of Adam-NSCL at GPM's energy threshold. On OPT-6.7b, it matches Adam-NSCL's forgetting at matched budget while learning more. As the theory predicts, the open/closed partition is the operative variable: open tokens beat random, sign-blind and anti-gate token sets on 18 of 18 seed-pairs. In pruning repair, every derivative-based local model of the output error at the dense weights is blind to pairs that open: the minimisers of the gate-weighted objective can leave the polyhedron, the objective's closed-form solution is 1.94 nats worse than no repair on OPT-1.3b, and a convex one-sided penalty bounds the escape.
