Daily AI Intake

1. 1. AREX: Towards a Recursively Self-Improving Agent for Deep Research

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 126 HF upvotes · 2 comments · 14 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Agent、Reasoning、Post-training/Alignment」。摘要显示它主要讨论 Deep research requires agents to find answers that jointly satisfy multiple constraints. Discovering such answers is costly, whereas verifying a candidate can often be decomposed into tractable constraint-wise checks. This discovery--verification asymmetry sug... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentReasoningPost-training/AlignmentEval/DataAI Infra
   PaperHFProjectGitHubPDF
   Original abstract

   Deep research requires agents to find answers that jointly satisfy multiple constraints. Discovering such answers is costly, whereas verifying a candidate can often be decomposed into tractable constraint-wise checks. This discovery--verification asymmetry suggests that a research agent should do more than simply search longer: it should recursively improve its current answer by verifying intermediate results and using the partially verified state to guide subsequent refinement. We introduce AREX, a family of Recursively Self-Improving (RSI) deep research agents. AREX alternates between an inner research loop that gathers evidence and constructs a provisional answer, and an outer self-improvement loop that audits the answer constraint-wise, identifies unresolved claims, and launches targeted follow-up research. To sustain RSI over long horizons, AREX learns an autonomous context-update tool that compresses growing interaction history into a compact improvement state preserving verified evidence and unresolved constraints, without relying on an external model. We train AREX on verified synthetic tasks and high-quality trajectories through agentic mid-training and long-horizon reinforcement learning. To mitigate sparse final rewards during long horizon learning, we emphasize key steps where decisive evidence is acquired or erroneous research directions are corrected. We instantia...

2. 2. K12-KGraph: A Curriculum-Aligned Knowledge Graph for Benchmarking and Training Educational LLMs

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 55 HF upvotes · 2 comments · 54 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「RAG/Memory、Reasoning、Multimodal」。摘要显示它主要讨论 Large language models are increasingly used in K-12 education, but existing benchmarks mainly test exam question answering rather than understanding how curriculum knowledge is structured and visually presented. We call this capability curriculum cognition. It... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察知识工作、企业搜索、长期记忆和本地资料库产品的新实现路径。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   RAG/MemoryReasoningMultimodalEval/DataAI4Science
   PaperHFProjectGitHubPDF
   Original abstract

   Large language models are increasingly used in K-12 education, but existing benchmarks mainly test exam question answering rather than understanding how curriculum knowledge is structured and visually presented. We call this capability curriculum cognition. It covers prerequisite chains, concept taxonomies, experiment-concept links, pedagogical sequencing, and visual grounding. We introduce K12-KGraph, a curriculum-aligned knowledge graph extracted from official People's Education Press textbooks in mathematics, physics, chemistry, and biology across primary, middle, and high school. It contains nine node types and fourteen relation types covering curriculum structure and visual grounding. From this graph, we derive K12-Bench, a 23,640-question multi-select benchmark with five task families: Ground, Prereq, Neighbor, Evidence, and Locate. We also build K12-Train, a graph-guided supervised fine-tuning corpus of 7,335 samples, including 2,267 text-only QA pairs and 5,068 multimodal VQA pairs. On K12-Bench, Gemini-3-Flash achieves only 57 percent exact match and Gemma-4-31B-IT reaches 46 percent, with Prereq and Neighbor being the hardest tasks. Our training experiments show that domain-specific supervision can reduce this gap. Under a matched 2,300-sample budget, K12-Train-Text consistently outperforms equally sized subsets of eight mainstream instruction-tuning corpora on Gaokao...

3. 3. ReferTrack: Referring Then Tracking for Embodied Visual Tracking

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 49 HF upvotes · 11 comments · 19 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Agent、Reasoning、Multimodal」。摘要显示它主要讨论 Embodied visual tracking (EVT) requires a mobile agent to continuously follow a specific target described in natural language using only onboard vision. While recent vision-language-action (VLA) policies unify target identification and trajectory planning, the... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentReasoningMultimodalEval/DataCode
   PaperHFProjectGitHubPDF
   Original abstract

   Embodied visual tracking (EVT) requires a mobile agent to continuously follow a specific target described in natural language using only onboard vision. While recent vision-language-action (VLA) policies unify target identification and trajectory planning, their chain-of-thought (CoT) reasoning often operates in abstract spatial latents that are difficult to supervise and weakly aligned with explicit image-space detections. To address this, we introduce ReferTrack, a referring-then-tracking paradigm that grounds EVT using a single forward-facing camera. Our model first selects the target from an indexed set of bounding boxes, then decodes tracking waypoints conditioned on this image-grounded decision. To preserve target motion cues over time, ReferTrack maintains a sliding-window queue of previously selected bounding boxes, injecting their geometric features into the visual history via temporal-viewpoint-bbox indicator (TVBI) tokens. We further enhance target identification by co-training on a custom Refer-QA dataset. On EVT-Bench, ReferTrack achieves state-of-the-art single-view performance with success rates of 89.4%, 73.3%, and 74.1% on the single-target, distracted, and ambiguity tracking splits, respectively -- matching or even surpassing several multi-camera baselines on identification-heavy tasks. Finally, real-world deployments on legged and humanoid robots validate its...

4. 4. Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 34 HF upvotes · 2 comments · 18 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Agent、Reasoning、Multimodal」。摘要显示它主要讨论 Spatial intelligence is essential for agents to move from static semantic understanding toward interacting with the physical world. Many spatial tasks are grounded in continuous visual scenes, where locations, regions, and paths are more naturally expressed by... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentReasoningMultimodalEval/Data
   PaperHFProjectGitHubPDF
   Original abstract

   Spatial intelligence is essential for agents to move from static semantic understanding toward interacting with the physical world. Many spatial tasks are grounded in continuous visual scenes, where locations, regions, and paths are more naturally expressed by pointing, marking, or drawing than by reporting precise coordinates or discrete textual symbols. Yet existing spatial reasoning benchmarks usually require coordinates, options, or text, creating an answer-interface mismatch for image-generation models. This makes it difficult to evaluate image-generation models under the same task semantics as text-output VLMs, despite their ability to externalize spatial judgments directly in pixel space. We propose ProVisE (Protocolized Visual Evaluation), a benchmark-agnostic framework that elicits protocol-constrained visual answers from image-generation models and parses them into structured predictions compatible with original metrics. ProVisE also includes an Agentic builder that constructs and validates task-specific protocols for new benchmarks. We further introduce SpatialGen-Bench, a curated diagnostic benchmark of 470 samples across 14 spatial subtasks, four capability levels, and diverse answer forms. We evaluate representative text-output VLMs and image-generation models in a unified setting and validate Agentic protocol construction on six external spatial benchmarks. Resul...

5. 5. Visual Contrastive Self-Distillation

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 41 HF upvotes · 2 comments · 6 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Reasoning、Multimodal、Eval/Data」。摘要显示它主要讨论 On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to ensure that the self-teacher provides a stronger learning sign... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察模型推理、规划、验证器和复杂任务能力是否有可复用技术路线。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   ReasoningMultimodalEval/DataAI InfraRobotics
   PaperHFProjectGitHubPDF
   Original abstract

   On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to ensure that the self-teacher provides a stronger learning signal than the student. Existing methods create this asymmetry either through privileged answers or visual evidence. We ask whether both can be removed, yielding a simpler form of OPSD driven purely by input conditioning. For this purpose, we propose Visual Contrastive Self-Distillation, namely VCSD, which converts image-content removal into an on-policy self-distillation signal. At each student-generated response prefix, the EMA teacher produces two next-token distributions under the same prompt and prefix -- one conditioned on the original image and the other on a content-erased control. Their token-wise log-probability difference highlights candidates whose likelihood is specifically increased by the instance-level visual content. We use this contrast to sharpen the teacher's original-image distribution within its plausible support, and distill the resulting full-distribution target into the student. Using ViRL39K dataset, VCSD consistently outperforms matched OPSD across Qwen3-VL and Qwen3.5 models. For example, on Qwen3-VL, it improves the seven-benchmark aggregate from 62.27% rightarrow 67.04% at 2B, 71.30% rightarrow 7...

6. 6. NVIDIA-labs OO Agents: Native Python Object-Oriented Agents

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 21 HF upvotes · 1 comments

   来自 Hugging Face Daily Papers，主题偏「Agent、RAG/Memory、Reasoning」。摘要显示它主要讨论 Traditional agent development is split across prompt templates, tool schemas, callback code, and workflow graphs. We present NVIDIA Object-Oriented Agents (NOOA), a model-agnostic Python framework for building reliable AI agents. NOOA takes a simpler approach:... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果标题正好贴近当前产品方向，值得点开原文看方法和实验设置；泛读时先存为观察项。

   AgentRAG/MemoryReasoningEval/DataCode
   PaperHFPDF
   Original abstract

   Traditional agent development is split across prompt templates, tool schemas, callback code, and workflow graphs. We present NVIDIA Object-Oriented Agents (NOOA), a model-agnostic Python framework for building reliable AI agents. NOOA takes a simpler approach: an agent is a Python object. Its methods are the actions the model can take, fields are its state, docstrings are its prompts, and its type annotations are contracts. A method whose code body consists of "..." is completed at runtime by an LLM-driven agent loop, while methods with normal bodies remain standard deterministic Python. This gives developers and agents the same interface, so agent behavior can be tested, traced, refactored, and improved just like other software. This paper makes three contributions. (1) We present the agent-as-a-Python-object programming model and the design principles behind it. Where Python has existing abstractions, we adopt them directly. Agent-specific capabilities--context, events, state rendering, long-term memory, and validated LLM loops--are exposed through simple Pythonic APIs, so both developers and agents share one familiar programming model. (2) We identify six model-facing ideas that NOOA is, to our knowledge, the first to combine on a single surface: typed input/output, pass-by-reference over live objects, code as action, programmable loop engineering, explicit object state, and...

7. 7. Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 20 HF upvotes · 1 comments

   来自 Hugging Face Daily Papers，主题偏「Agent、RAG/Memory、Multimodal」。摘要显示它主要讨论 We introduce Tencent WorkBuddy Bench, a multi-domain evaluation suite for coding agents; this report documents its construction methodology, scoring protocol, and a cross-model leaderboard. At its core is a unified evaluation framework for constructing and run... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentRAG/MemoryMultimodalEval/DataCode
   PaperHFProjectPDF
   Original abstract

   We introduce Tencent WorkBuddy Bench, a multi-domain evaluation suite for coding agents; this report documents its construction methodology, scoring protocol, and a cross-model leaderboard. At its core is a unified evaluation framework for constructing and running distribution-informed coding-agent tasks across four work domains - Code, Web, Office, and Security. Rather than adapting public issue text, every task is reverse-engineered from a real commit, pull request, or business scenario and rewritten as a short, colloquial, role-played request, so that a task's prompt is not recoverable by web-searching the underlying issue, pull request, or commit thread. Because the dataset is released openly - task directories, environment images, evaluation harness, tests, and reference solutions - contamination resistance rests on this construction together with dataset versioning rather than on secrecy. The four subsets - repository-level engineering, front-end development, office and business workflows, and red-/blue-team security - probe complementary facets of real work, each with its own verification style. All are packaged in a uniform task-directory format and run, under a uniform and reproducible protocol, on two agent harnesses (CodeBuddy Code and Claude Code); the full open release makes the benchmark reproducible end to end and directly auditable, since any third party can re-...

8. 8. LLMs Get Lost in Evolving User Intent

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 18 HF upvotes · 4 comments · 2 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Agent、Eval/Data、AI Infra」。摘要显示它主要讨论 As LLMs become more capable, they are increasingly deployed as collaborative agents, taking on user-delegated tasks through iterative interaction. Yet genuine interaction is inherently dynamic: users rarely specify their intent upfront, instead disclosing, rev... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentEval/DataAI InfraRobotics
   PaperHFGitHubPDF
   Original abstract

   As LLMs become more capable, they are increasingly deployed as collaborative agents, taking on user-delegated tasks through iterative interaction. Yet genuine interaction is inherently dynamic: users rarely specify their intent upfront, instead disclosing, revising, and reshaping it as the conversation unfolds. Despite this, LLMs are still predominantly evaluated or trained in single-turn, fully-specified settings, leaving open a fundamental question: how well do LLMs track and act on user intent as it evolves over the course of a conversation? To study this, we introduce a framework that transforms static, single-turn tasks into dynamic multi-turn conversations in which the user's intent evolves across turns--incrementally revealed, revised, and at times redirected mid-conversation--while preserving each task's original evaluation protocol, enabling existing benchmarks to be reused as controlled testbeds without new annotation. Across multiple tasks, we surface a consistent phenomenon: strong static-setting performance does not transfer to the evolving-intent setting, with substantial drops across model families. Our findings point to a fundamental gap: today's LLMs do not yet faithfully track and act on the user's evolving intent, a capability invisible to static evaluation yet critical for future collaborative agents.

9. 9. Multi-Turn On-Policy Distillation with Prefix Replay

   Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 8 HF upvotes · 2 comments · 4 GitHub stars

   来自 Hugging Face Daily Papers，主题偏「Agent、Reasoning、AI Infra」。摘要显示它主要讨论 We study on-policy distillation (OPD) for agentic tasks, where an LLM agent interacts with an environment over multiple turns and a student imitates a teacher over these multi-turn interaction histories. Fully online OPD is costly because each update requires... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

   为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

   读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

   AgentReasoningAI Infra
   PaperHFProjectGitHubPDF
   Original abstract

   We study on-policy distillation (OPD) for agentic tasks, where an LLM agent interacts with an environment over multiple turns and a student imitates a teacher over these multi-turn interaction histories. Fully online OPD is costly because each update requires fresh student rollouts through the environment and teacher queries at visited histories. We propose Replayed-Prefix On-Policy Distillation (ReOPD), an off-environment alternative that reuses pre-collected teacher trajectories as replayed prefixes: the student acts at selected steps, while the teacher provides dense per-step supervision without executing new environment interactions. We show that multi-turn OPD introduces a prefix trap: making histories more student-on-policy improves relevance to the student, but can query the teacher on histories where its target is unreliable. This creates a two-sided distribution shift between student occupancy and teacher reliability. ReOPD addresses this by treating multi-turn OPD as a reliability-aware prefix distribution design and implements it with a simple step-decaying sampling schedule that emphasizes early, lower-shift prefixes. Across mathematical reasoning with Python and search environments over multiple teacher and student model scales, ReOPD preserves or improves OPD-level accuracy, uses zero tool calls during student training, and is at least 4times faster per rollout th...

10. 10. FinanceComplexQA: Benchmarking Agentic Reasoning on Industrial-grade Financial Documents

    Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 5 HF upvotes · 2 comments · 5 GitHub stars

    来自 Hugging Face Daily Papers，主题偏「Agent、RAG/Memory、Reasoning」。摘要显示它主要讨论 Agentic Reasoning has become a transformative force in financial analysis due to its ability to integrate large-scale information and generate reliable and accurate content. However, when handling complex real-world problems, different agents still show signif... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

    为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

    读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

    AgentRAG/MemoryReasoningEval/Data
    PaperHFProjectGitHubPDF
    Original abstract

    Agentic Reasoning has become a transformative force in financial analysis due to its ability to integrate large-scale information and generate reliable and accurate content. However, when handling complex real-world problems, different agents still show significant performance variation. In this work, we design Finance-LaTeX SKILL, a skill for synthesizing financial documents with complex layouts based on expert knowledge. Using an agent workflow built on this skill, we generate 2,000 professional financial documents along with 6,000 high-quality question-answer pairs. To evaluate the overall capability of agents, we introduce FinanceComplexQA, a comprehensive open-ended generation benchmark for financial documents that closely resembles real-world scenarios. It contains 2,026 deep research tasks targeting 1009 financial documents. FinanceComplexQA has 8 key features: bilingual support; coverage of six mainstream scenarios and seven tasks; expert-level document reasoning questions; deep research of complex layouts; relatively stable and permanent reference answers; and precise evaluation through an Agent-as-a-Judge with multiple evaluation metrics. Using FinanceComplexQA, we conduct a comprehensive evaluation of leading RAG systems and agentic reasoning tools for financial document QA. Through identifying and analyzing failure cases, we provide an in-depth study of their capabi...

11. 11. OpenForgeRL: Train Harness-native Agents in Any Environment

    Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 5 HF upvotes · 3 comments

    来自 Hugging Face Daily Papers，主题偏「Agent、RAG/Memory、Reasoning」。摘要显示它主要讨论 Modern AI agents rely on elaborate inference harnesses such as Claude Code, Codex, and OpenClaw to drive multi-turn reasoning, tool use, and access to external systems. While powerful, these complex harnesses also make agents hard to train end-to-end with open... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

    为什么值得看适合观察 agentic RL、工具调用、工作流自动化或软件代理能力是否出现新方法。

    读原文判断如果标题正好贴近当前产品方向，值得点开原文看方法和实验设置；泛读时先存为观察项。

    AgentRAG/MemoryReasoningMultimodalEval/Data
    PaperHFPDF
    Original abstract

    Modern AI agents rely on elaborate inference harnesses such as Claude Code, Codex, and OpenClaw to drive multi-turn reasoning, tool use, and access to external systems. While powerful, these complex harnesses also make agents hard to train end-to-end with open infrastructure, whose SFT/RL stacks cannot natively express stateful, multi-process harness inference. To address this, we present OpenForgeRL, an open-source framework for training harness-based agents end-to-end in diverse environments. OpenForgeRL achieves this with a lightweight proxy that serves the harness's model calls while recording them as training data for a standard RL codebase (e.g., veRL), and a Kubernetes orchestrator that runs each rollout in its own remote container, together enabling training on any harness in any environment at scale. By decoupling training and inference, OpenForgeRL allows researchers to easily train, study, and improve agents directly in the real harnesses and environments they are deployed with. We validate our framework across diverse, complex harnesses and environments, spanning tool/claw-based agents and multimodal GUI browser- and computer-use agents. Using only hundreds to a few thousand tasks, OpenForgeClaw reaches 31.7 pass^3 and 55.9 pass@3 on ClawEval and 33.7 on QwenClawBench. OpenForgeGUI reaches 37.7 on OSWorld-Verified, 63.0 on Online-Mind2Web, and 72.3 on WebVoyager. Bo...

12. 12. Dataset Distillation by Influence Matching

    Hugging Face Daily Papers 3 天前 · 2026/07/24 08:00 1 HF upvotes · 2 comments

    来自 Hugging Face Daily Papers，主题偏「RAG/Memory、Multimodal、Post-training/Alignment」。摘要显示它主要讨论 We revisit dataset distillation from an outcome-centric perspective. Rather than aligning process surrogates (per-step gradients or training trajectories), Influence Matching (Inf-Match) aligns the final outcome of training: it learns a compact synthetic set w... 先把它当作时效信号看：判断它是否正在影响 agent、RAG、多模态、post-training、评测或 AI infra 的产品/研究方向。

    为什么值得看适合观察知识工作、企业搜索、长期记忆和本地资料库产品的新实现路径。

    读原文判断如果你要找可复现 demo、开源工具或产品化线索，建议点开项目/GitHub；否则先看中文摘要即可。

    RAG/MemoryMultimodalPost-training/AlignmentEval/DataAI Infra
    PaperHFProjectGitHubPDF
    Original abstract

    We revisit dataset distillation from an outcome-centric perspective. Rather than aligning process surrogates (per-step gradients or training trajectories), Influence Matching (Inf-Match) aligns the final outcome of training: it learns a compact synthetic set whose effect on the converged parameters matches that of the full dataset. Concretely, we introduce a fully differentiable, sample-level influence estimator that quantifies parameter shifts from adding or removing data, without time-consuming inverse-Hessian products or convexity assumptions. The estimator runs in linear time by unrolling the optimization dynamics and applying a first-order Taylor approximation. We then learn the synthetic set by minimizing the mismatch between its influence and that of the real dataset, yielding outcome alignment rather than heuristic process imitation. Inf-Match delivers the best accuracy across standard classification benchmarks. For instance, on Tiny-ImageNet (IPC=10), Inf-Match attains 31.5\%, a +4.7\% improvement over NCFM. Beyond classification, Inf-Match scales to vision-language distillation on Flickr30K, outperforming strong process-matching baselines. For instance, with 200 to 1000 synthetic samples, our method achieved a leading impressive average on image/text retrieval tasks, higher than NCFM by 2.5\%. The code will be released via https://github.com/hrtan/infmatch.

Using cached daily papers because the latest fetch failed.