Scoring Rules! Statistical and Strategic Alignment for Text Evaluation Metrics

Reference-based text evaluation metrics, which are widely used to assess natural language generation systems, score a candidate response by comparing it with a reference response. The reliability of an evaluation metric is usually judged by its statistical correlation with human ratings. However, as these metrics are increasingly used as optimization objectives, correlation alone is no longer sufficient: agents may strategically game the evaluation metric. We study this issue through two complementary notions of alignment. A metric is statistically aligned if it correlates with human ratings and strategically aligned if it resists perturbations that do not add task-relevant information. We make two contributions. First, we propose test principles for reference-based metrics consisting of human-rating correlation, degradation sensitivity, and manipulation robustness. These principles evaluate whether a metric agrees with human judgments, penalizes low-effort information loss, and resists strategic score inflation. Second, we develop a unified design framework for mutual-information-based metrics that decomposes existing and new metrics into four choices: information measure, estimation method, text representation, and prediction mechanism. Across peer review, summarization, and question answering, we find that strong human-rating correlation does not imply strategic alignment: LLM-as-a-Judge achieves high correlation but is susceptible to manipulation. In contrast, mutual-information-based metrics substantially improve manipulation robustness. Our framework also uncovers a new metric that achieves the strongest overall robustness in our experiments while remaining competitive on human-rating correlation.

OnlineCache: Learning Dynamic Caching Policies with Error Correction for Efficient Diffusion Inference

Diffusion models have revolutionized generative tasks but incur high latency due to iterative denoising. While cache-based strategies accelerate inference by reusing intermediate features, they largely rely on static, sample-agnostic schedules. We argue that this rigidity overlooks two facts empirically validated in this paper: (i) generation difficulty varies across prompts, requiring adaptive resource allocation--complex inputs demand more computation while simpler ones require less; (ii) error sensitivity fluctuates across timesteps, where static policies may cache high-error steps or waste computation on low-error ones. We therefore propose OnlineCache, a dynamic caching framework that jointly learns when to cache and how to correct approximation errors. We leverage policy gradient to train a lightweight network for adaptive speed-quality trade-offs, and incorporate a learnable corrector to mitigate caching-induced errors. Both modules are jointly optimized under a bilevel optimization framework, with the policy targeting global generation quality and the corrector minimizing local errors. Our method automatically allocates computational resources across both samples and timesteps, improving overall generation quality. Extensive experiments demonstrate clear superiority. On FLUX.1-dev model, OnlineCache achieves nearly 3 speedup while preserving generation fidelity. On DiT and CogVideoX, it similarly delivers competitive acceleration without compromising quality; across all scenarios, it consistently outperforms existing cache-based acceleration baselines.

Tensor-Based Joint Pitch and DOA Estimation

We propose a tensor-based subspace method for the joint estimation of pitch and direction of arrival (DOA) of multiple harmonic sources. Unlike conventional matrix-based approaches that unfold the spatio-temporal data into a single matrix, the proposed method preserves the multidimensional Hankel tensor structure and exploits a Kronecker product constraint satisfied by the true signal subspace. The matrix-based subspace estimate is refined by projecting it onto an empirically estimated Kronecker-structured signal cage obtained from mode-unfolded tensor subspaces. Beyond the estimator itself, we provide a non-asymptotic theory explaining when this tensor refinement improves the matrix baseline. We show that the oracle refinement using the true Kronecker projector is non-worsening, decompose the overall tensor gain into oracle gain and empirical loss, and derive deterministic and high-probability sufficient conditions for positive overall tensor gain under complex Gaussian noise. The analysis reveals a sweet-spot behavior: the largest certified gain occurs when the matrix estimate is neither nearly perfect nor too inaccurate. Simulations over closely spaced, harmonically related, and coherent-source scenarios show that the proposed method improves subspace accuracy and downstream pitch/DOA estimation, and also outperforms representative matrix and tensor-train-based baselines.

HalluScope: Fine-grained Hallucination Diagnosis for Multimodal Large Language Models

Although Multimodal Large Language Models have achieved strong performance across a wide range of vision-language tasks, they still suffer from hallucinations, where model outputs become inconsistent with the visual content, textual context, or commonsense knowledge. Existing studies primarily address this problem through coarse-grained detection. However, these approaches often provide insufficient diagnostic information for understanding hallucination types and supporting downstream hallucination mitigation. To bridge this gap, we propose fine-grained hallucination diagnosis for MLLMs, a new unified task that jointly performs hallucination detection, classification, and interpretable explanation generation. We develop an automated data generation pipeline and construct HalluScope-30K, a large-scale diagnostic dataset covering eight sources and five task categories. Based on this dataset, we design a multi-granular joint reward function and train two diagnosis models, HalluScope-4B and HalluScope-8B, which achieve state-of-the-art performance on both the MHALO benchmark and our fine-grained hallucination classification benchmark. Notably, detection and classification are mutually beneficial under joint optimization. Furthermore, diagnosis-driven feedback experiments show that the fine-grained diagnostic explanations produced by our model effectively guide target models to correct their hallucinations, with full diagnosis substantially outperforming all baselines on both Qwen3-VL-8B-Instruct and LLaVA-1.5-7B.

Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents

In long-horizon tasks, decision-relevant state is often scattered across an expanding trajectory, while the action agent must surface it and act. As trajectories grow, task requirements, environment facts, prior attempts, diagnoses, and open subgoals can be buried in the context window or pushed beyond it, failing to influence decisions when needed. We call this failure mode "behavioral state decay". We study memory as an active intervention mechanism rather than passive retrieval. A separate memory agent runs alongside an unmodified action agent, updating a structured memory bank from the recent trajectory and deciding whether to inject a memory-grounded reminder or remain silent. The module is plug-and-play with frontier action agents and existing agent harnesses. Across Terminal-Bench 2.0 and $τ^2$-Bench, it improves pass@1 for both weaker and stronger action agents, with gains of +8.3 pp on Terminal-Bench and +6.8 pp on $τ^2$-Bench. Ablations show that selective intervention outperforms passive bank exposure, always-on injection, advisor-only guidance, and general retrieval. As an early step toward open-weight memory policies, we train Qwen3.5-27B on SETA using SFT and GRPO, improving validation reward and achieving partial transfer to Terminal-Bench.

UI2App: Benchmarking Visual Interaction Inference in Executable Web Application Generation

Large language models (LLMs) have demonstrated growing competence in web page generation. However, existing text-driven approaches rely on complex prompts that impose substantial demands on users and offer limited expressivity for page layout and cross-page visual coherence. Image-driven paradigms, which take UI screenshots as input, align more closely with real development workflows. However, current benchmarks focus primarily on visual fidelity and lack a systematic evaluation of the interaction capabilities in generated artifacts. To address this gap, we introduce UI2App, the first benchmark targeting interaction inference, the ability to recover application behavior from screenshots alone, without any textual or behavioral guidance. UI2App comprises 327 screenshots grouped into 45 state-coherent screenshot sets for runnable multi-route web applications. We design an end-to-end pipeline that evaluates each artifact along four dimensions: executability, navigation reachability, visual fidelity, and interaction inference. The interaction metric (IIS) assesses inferred interactions by functional correctness and state-management complexity, crediting any valid implementation rather than matching a single reference. Experiments on six frontier vision-language models reveal a marked capability mismatch between visual reconstruction and interaction realization: the visual-fidelity leader scores only 7.5 on IIS, ranking fourth and trailing the IIS leader by 5.2x. High-complexity interactions such as cross-page state remain a pervasive bottleneck, with half of the evaluated models scoring exactly zero on this dimension. Overall, the results indicate that inferring complete interaction behavior from static screenshots remains a key challenge for models.

Aligning Language Models with Selective Prediction

Large language models (LLMs) are increasingly deployed as critical decision-making components in high-stakes real-world AI systems, rendering LLM reliability a foremost practical concern. In this paper, we focus on enhancing LLM reliability through selective prediction (SP), a strategy that allows an LLM to only predict for inputs where it is likely to be correct (i.e., coverage) and hence reduce the error rate (i.e., risk) on that portion of inputs -- flagging the remaining inputs for future human discretion. In other words, SP improves LLM reliability by balancing the risk-coverage trade-off and enabling seamless human-AI collaboration. To integrate SP into LLMs, we focus on the LLM post-training alignment stage and propose to align LLMs with SP performance metrics, in contrast with existing LLM alignment methods that focus primarily on correctness or calibration metrics. Specifically, we propose a novel alignment framework, Reinforcement Learning for Selection Reward (RLSR), which targets the area under the risk-coverage curve (AURC) -- a popular SP performance metric -- as its alignment objective. RLSR achieves substantially better risk-coverage trade-off compared to multiple alignment baselines on both in-domain and out-of-domain tasks.

SWE-Together: Evaluating Coding Agents in Interactive User Sessions

Most coding-agent benchmarks are static: an agent receives a complete task description up front and is judged only by its final code. Real coding assistance is interactive, with users clarifying goals, adding constraints, and correcting mistakes over multiple turns. We introduce SWE-Together, a multi-turn benchmark reconstructed from real user-agent coding sessions. To make real interactions verifiable, we curate 109 repository-level tasks from 11,260 recorded sessions, selecting sessions with recoverable repository states, clear user goals, and observable outcomes. To replay these interactions across agents, we build a reactive LLM-based user simulator that preserves the original users' intents and provides feedback when the coding agent's progress requires it. To evaluate agents as collaborators, we measure both final repository correctness and the number of corrective feedback turns required during the interaction. Experiments with frontier coding agents show that stronger agents generally achieve higher final success rates while requiring fewer interventions, suggesting an improved user experience.

Distill on a Diet: Efficient Knowledge Distillation via Learnable Data Pruning

Knowledge Distillation (KD) is widely used to obtain compact models for efficient inference in resource-constrained environments. Yet the computational overhead of the distillation process itself is often overlooked, raising the question of whether a better student model can be obtained with less data and less compute via data pruning. However, existing data pruning methods are not designed for KD: some introduce substantial overhead, such as obtaining training dynamics through retraining, while others rely on heuristic selection rules that fail to capture what KD actually requires, often resulting in suboptimal subsets. To address these issues, we propose IF-Beta, an efficient data pruning framework that combines influence functions with a learnable sampling policy. Empirically, we first demonstrate that influence functions can serve as an effective and efficient estimator of sample impact in KD settings, where only a pretrained teacher is available. Building on this, our sampling policy is specifically parameterized by a Beta distribution, whose highly flexible two-parameter family allows the policy to adapt to diverse pruning regimes rather than being tied to fixed heuristic forms. Next, we formulate KD pruning as optimizing this policy through a bilevel objective, where the inner loop operates in the teacher feature space with a KD-aligned objective, enabling fast proxy training, while the outer loop updates the policy parameters to maximize distillation performance. This design ensures that IF-Beta is both computationally efficient and inherently aligned with the goals of KD. Extensive experiments on CIFAR-10/100 and ImageNet show that IF-Beta consistently outperforms other baselines across a wide range of pruning ratios. Remarkably, IF-Beta enables students trained on less data and less compute to surpass the performance of students distilled on the full dataset.

VQActFlow: Vector-Quantized Action Mode Steering for Multi-Task Robot Manipulation

Multi-task robot manipulation policies are challenging to learn from demonstration because traditionally a single network must select among qualitatively different action modes from a multimodal demonstration distribution, conditioned on language and visual context. A wrong mode selection means executing the wrong task or an action infeasible in the scene. Tokenizing continuous actions into a learned discrete codebook separates these modes at the representation level, offering structural advantages for multi-task learning. We propose VQActFlow, a multi-task manipulation policy that tokenizes action chunks and generates code sequences via Variational Flow Matching. VQActFlow maintains an explicit preference over action modes throughout generation. Inference-time guidance acts on this preference to steer mode commitment. We instantiate this with classifier-free guidance over language conditioning, which steers the policy toward the instructed action mode, and a learned codebook critic that supplies a complementary feasibility signal. We evaluate VQActFlow on three platforms: the LIBERO simulation benchmarks, a Unitree G1 humanoid performing whole-body pick-and-place, and an ALOHA-style bimanual platform performing contact-rich tasks. Across these benchmarks, VQActFlow outperforms both continuous and discrete baselines.