AI ethics on Thursday, 19 March 2026

Open-Vocabulary Semantic Segmentation (OVSS) assigns pixel-level labels from an open set of text-defined categories, demanding reliable generalization to unseen classes at inference. Although modern vision-language models (VLMs) support strong open-vocabulary recognition, their representations learned through global contrastive objectives remain suboptimal for dense prediction, prompting many OVSS methods to depend on limited adaptation or refinement of image-text similarity maps. This, in turn,

This paper computationally investigates whether thought requires a language-like format, as posited by the Language of Thought (LoT) hypothesis. We introduce the ``AI Private Language'' thought experiment: if two artificial agents develop an efficient, inscrutable communication protocol via multi-agent reinforcement learning (MARL), and their performance declines when forced to use a human-comprehensible language, this Efficiency Attenuation Phenomenon (EAP) challenges the LoT. We formalize this

Large language models (LLMs) are increasingly deployed as agents with access to executable tools, enabling direct interaction with external systems. However, most safety evaluations remain text-centric and assume that compliant language implies safe behavior, an assumption that becomes unreliable once models are allowed to act. In this work, we empirically examine how executable tool affordance alters safety alignment in LLM agents using a paired evaluation framework that compares text-only chat

Most AI literacy courses for non-technical undergraduates emphasize conceptual breadth over technical depth. This paper describes UNIV 182, a prerequisite-free course at George Mason University that teaches undergraduates across majors to understand, use, evaluate, and build AI systems. The course is organized around five mechanisms: (1) a unifying conceptual pipeline (problem definition, data, model selection, evaluation, reflection) traversed repeatedly at increasing sophistication; (2) concur

Recent vision-language models (VLMs) have shown strong generalization and multimodal reasoning abilities in natural domains. However, their application to medical diagnosis remains limited by the lack of comprehensive and structured datasets that capture real clinical workflows. To advance the development of VLMs for clinical applications, particularly in gastric cancer, we introduce Gastric-X, a large-scale multimodal benchmark for gastric cancer analysis providing 1.7K cases. Each case in Gast

This work examines how AI, especially agentic systems, is being adopted in engineering and manufacturing workflows, what value it provides today, and what is needed for broader deployment. This is an exploratory and qualitative state-of-practice study grounded in over 30 interviews across four stakeholder groups (large enterprises, small/medium firms, AI developers, and CAD/CAM/CAE vendors). We find that near-term AI gains cluster around structured, repetitive work and data-intensive synthesis,

Most social choice rules assume access to full rankings, while current alignment practice -- despite aiming for diversity -- typically treats voters as anonymous and comparisons as independent, effectively extracting only about one bit per voter. Motivated by this gap, we study social choice under an extreme communication budget in the linear social choice model, where each voter's utility is the inner product between a latent voter type and the embedding of the context and candidate. The candid

High-quality GPS trajectories are essential for location-based web services and smart city applications, including navigation, ride-sharing and delivery. However, due to low sampling rates and limited infrastructure coverage during data collection, real-world trajectories are often sparse and feature unevenly distributed location points. Recovering these trajectories into dense and continuous forms is essential but challenging, given their complex and irregular spatio-temporal patterns. In this

Security in LLM agents is inherently contextual. For example, the same action taken by an agent may represent legitimate behavior or a security violation depending on whose instruction led to the action, what objective is being pursued, and whether the action serves that objective. However, existing definitions of security attacks against LLM agents often fail to capture this contextual nature. As a result, defenses face a fundamental utility-security tradeoff: applying defenses uniformly across

Self-improving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problem-solving processes. Existing approaches to self-improvement rely on fixed, handcrafted meta-level mechanisms, fundamentally limiting how fast such systems can improve. The Darwin Gödel Machine (DGM) demonstrates open-ended self-improvement in coding by repeatedly generating and evaluating self-modified variants. Because both evaluation and self-modification are coding tasks,

Large language model (LLM) agents increasingly rely on external tools (file operations, API calls, database transactions) to autonomously complete complex multi-step tasks. Practitioners deploy defense-trained models to protect against prompt injection attacks that manipulate agent behavior through malicious observations or retrieved content. We reveal a fundamental \textbf{capability-alignment paradox}: defense training designed to improve safety systematically destroys agent competence while f

Real-world financial decision-making is a challenging problem that requires reasoning over heterogeneous signals, including company fundamentals derived from regulatory filings and trading signals computed from price dynamics. Recently, with advances in Large Language Models (LLMs), financial analysts have begun to use them for financial decision-making tasks. However, existing financial question-answering benchmarks for testing these models primarily focus on company balance sheet data and rare

Understanding and generating 3D objects as compositions of meaningful parts is fundamental to human perception and reasoning. However, most text-to-3D methods overlook the semantic and functional structure of parts. While recent part-aware approaches introduce decomposition, they remain largely geometry-focused, lacking semantic grounding and failing to model how parts align with textual descriptions or their inter-part relations. We propose DreamPartGen, a framework for semantically grounded, p

Conventional pixel-wise loss functions fail to enforce topological constraints in coronary vessel segmentation, producing fragmented vascular trees despite high pixel-level accuracy. We present ARIADNE, a two-stage framework coupling preference-aligned perception with RL-based diagnostic reasoning for topologically coherent stenosis detection. The perception module employs DPO to fine-tune the Sa2VA vision-language foundation model using Betti number constraints as preference signals, aligning t

Large language models frequently exhibit suboptimal performance on low resource languages, primarily due to inefficient subword segmentation and systemic training data imbalances. In this paper, we propose Variable Entropy Policy Optimization (VEPO), which leverages Reinforcement Learning with Verifiable Rewards to incorporate deterministic structural constraints into the policy alignment process. This framework ensures prescribed sequence length, robust format consistency, and rigorous linguist

Large language models (LLMs) exhibit pronounced social biases. Output-level or data-optimization--based debiasing methods cannot fully resolve these biases, and many prior works have shown that biases are embedded in internal representations. We propose \underline{U}nified \underline{G}raph \underline{I}somorphism for \underline{D}ebiasing large language models (\textit{\textbf{UGID}}), an internal-representation--level debiasing framework for large language models that models the Transformer as

The creation of high-fidelity, customizable 3D indoor scene textures remains a significant challenge. While text-driven methods offer flexibility, they lack the precision for fine-grained, instance-level control, and often produce textures with insufficient quality, artifacts, and baked-in shading. To overcome these limitations, we introduce CustomTex, a novel framework for instance-level, high-fidelity scene texturing driven by reference images. CustomTex takes an untextured 3D scene and a set

Generative AI (GenAI) is increasingly being integrated into the online ecosystem, including online health communities (OHCs), where people with diverse health conditions exchange social support. For example, in OHCs, support providers are beginning to share content generated, directly or indirectly, by popular GenAI-based tools. OHCs are governed by norms that define appropriate behavior when providing support. Ways in which AI-generated support interacts with these norms remain underexplored. I

Existing cybersecurity literature lacks a source of empirical, representative data as to the true nature of cyberattacks on Critical National Infrastructure. We have obtained UK-wide data on incidents reported under the Network and Information Systems (NIS) Regulations in 2024 causing "a significant impact on the continuity" of essential services and comparator data from intelligence agencies. We find that 29% of NIS reports already concern cybersecurity incidents. As the UK Government seeks to

Four-term word analogies (A:B::C:D) are classically modeled geometrically as ''parallelograms,'' yet recent work suggests 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) anal

The integration of artificial intelligence (AI) technologies into judicial decision-making, particularly in pretrial, sentencing, and parole contexts, has generated substantial concerns about transparency, reliability, and accountability. At the same time, these developments have brought the limitations of human judgment into sharper relief and underscored the importance of understanding how judges interact with AI-based decision aids. Using criminal justice risk assessment as a focal case, we c

Models that bridge vision and language, such as CLIP, are key components of multimodal AI, yet their large-scale, uncurated training data introduce severe social and spurious biases. Existing post-hoc debiasing methods often operate directly in the dense CLIP embedding space, where bias and task-relevant information are highly entangled. This entanglement limits their ability to remove bias without degrading semantic fidelity. In this work, we propose Sparse Embedding Modulation (SEM), a post-ho

This study investigates the adoption and effectiveness of AI-based anomaly detection in cross-provider electronic health record (EHR) environments. It aims to (1) identify the organisational and digital capabilities required for successful implementation and (2) evaluate the performance and interpretability of lightweight anomaly detection approaches using contextual audit data. A semi-systematic scoping synthesis is conducted to derive a four-pillar readiness framework covering governance, infr

Retrieval-Augmented Generation (RAG) improves Large Language Models (LLMs) by grounding generation in external, non-parametric knowledge. However, when a task requires choosing among competing options, simply grounding generation in broadly relevant context is often insufficient to drive the final decision. Existing RAG methods typically rely on a single initial query, which often favors topical relevance over decision-relevant evidence, and therefore retrieves background information that can fa

We study online resource allocation among N interacting modules over T rounds. Unlike standard online optimization, costs are endogenous: they depend on the full allocation vector through an interaction matrix W encoding pairwise cooperation and competition. We analyze three paradigms: (I) uniform allocation (cost-ignorant), (II) gated allocation (cost-estimating), and (III) competitive allocation via multiplicative weights update with interaction feedback (cost-revealing). Our main results esta

Natural language prompts often suffer from intent transmission loss: the gap between what users actually need and what they communicate to AI systems. We evaluate PPS (Prompt Protocol Specification), a 5W3H-based framework for structured intent representation in human-AI interaction. In a controlled three-condition study across 60 tasks in three domains (business, technical, and travel), three large language models (DeepSeek-V3, Qwen-Max, and Kimi), and three prompt conditions - (A) simple promp

Generative AI services like ChatGPT and Gemini are some of the fastest-growing consumer services. Individuals using such services must accept their terms of use before access, and conform to these terms for continued use of the service. Established literature has shown that despite their status as legally-binding agreements, terms of use are not actually well-understood, and may contain implications that are surprising for consumers. In this paper, we analyse the terms of 6 generative AI service

This Data Descriptor presents a fully open, multi-modal dataset for estimating vertical ground reaction force (vGRF) from consumer-grade Apple Watch sensors with laboratory force plate ground truth. Ten healthy adults aged 26--41 years performed five activities: walking, jogging, running, heel drops, and step drops, while wearing two Apple Watches positioned at the left wrist and waist. The dataset contains 492 validated trials with time-aligned inertial measurement unit (IMU) recordings (approx

This paper presents a manifesto that articulates the conceptual foundations of Agentic Business Process Management (APM), an extension of Business Process Management (BPM) for governing autonomous agents executing processes in organizations. From a management perspective, APM represents a paradigm shift from the traditional process view of the business process, driven by the realization of process awareness and an agent-oriented abstraction, where software and human agents act as primary functio

The rapid proliferation of artificial intelligence (AI) technologies has led to a dynamic regulatory landscape, where legislative frameworks strive to keep pace with technical advancements. As AI paradigms shift towards greater autonomy, specifically in the form of agentic AI, it becomes increasingly challenging to precisely articulate regulatory stipulations. This challenge is even more acute in the domains of security and privacy, where the capabilities of autonomous agents often blur traditio

Knowledge-grounded dialogue systems aim to generate informative, contextually relevant responses by conditioning on external knowledge sources. However, most existing approaches focus exclusively on English, lack explicit citation mechanisms for verifying factual claims, and offer limited transparency into model decision-making. We present XKD-Dial, a progressive four-stage training pipeline for explainable, knowledge-grounded dialogue generation in a bilingual (English-Hindi) setting, comprisin

Language models increasingly appear to learn similar representations, despite differences in training objectives, architectures, and data modalities. This emerging compatibility between independently trained models introduces new opportunities for cross-model alignment to downstream objectives. Moreover, this capability unlocks new potential application domains, such as settings where security, privacy, or competitive constraints prohibit direct data or model sharing. In this work, we investigat

Artificial intelligence (AI) systems are deployed as collaborators in human decision-making. Yet, evaluation practices focus primarily on model accuracy rather than whether human-AI teams are prepared to collaborate safely and effectively. Empirical evidence shows that many failures arise from miscalibrated reliance, including overuse when AI is wrong and underuse when it is helpful. This paper proposes a measurement framework for evaluating human-AI decision-making centered on team readiness. W

Large language models are increasingly proposed as autonomous agents for high-stakes public workflows, yet we lack systematic evidence about whether they would follow institutional rules when granted authority. We present evidence that integrity in institutional AI should be treated as a pre-deployment requirement rather than a post-deployment assumption. We evaluate multi-agent governance simulations in which agents occupy formal governmental roles under different authority structures, and we s

Autonomous agents can produce harmful behavioral patterns from individually valid requests -- a threat class per-request policy evaluation cannot address, because stateless engines evaluate each request in isolation. We present ACP, a temporal admission control protocol enforcing behavioral properties over execution traces via static risk scoring combined with stateful signals (anomaly accumulation, cooldown) through a LedgerQuerier abstraction. ACP blocks execution based on deterministic, histo

An investigation, from a gender perspective, of how students view the ethical implications and societal effects of artificial intelligence is conducted, examining concepts that could have a big influence on how artificial intelligence may be taught in the future. For this, we conducted a survey on a cohort of 230 second year computer science students to reveal their opinions. The results revealed that AI, from the students' perspective, will significantly impact daily life, particularly in areas

Diffusion Large Language Models (dLLMs) introduce a new paradigm for language generation, which in turn presents new challenges for aligning them with human preferences. In this work, we aim to improve the policy optimization for dLLMs by reducing the cost of the trajectory probability calculation, thereby enabling scaled-up offline policy training. We prove that: (i) under reference policy regularization, the probability ratio of the newly unmasked tokens is an unbiased estimate of that of inte

Standard attention mechanisms in transformers are limited by their pairwise formulation, which hinders the modeling of higher-order dependencies among tokens. We introduce the NeuroGame Transformer (NGT) to overcome this by reconceptualizing attention through a dual perspective: tokens are treated simultaneously as players in a cooperative game and as interacting spins in a statistical physics system. Token importance is quantified using two complementary game-theoretic concepts -- Shapley value

Flow Matching (FM) has recently emerged as a leading approach for high-fidelity visual generation, offering a robust continuous-time alternative to ordinary differential equation (ODE) based models. However, despite their success, FM models are highly sensitive to dataset biases, which cause severe semantic degradation when generating out-of-distribution or minority-class samples. In this paper, we provide a rigorous mathematical formalization of the ``Bias Manifold'' within the FM framework. We

Natural language explanations provide an inherently human-understandable way to explain black-box models, closely reflecting how radiologists convey their diagnoses in textual reports. Most works explicitly supervise the explanation generation process using datasets annotated with explanations. Thus, though plausible, the generated explanations are not faithful to the model's reasoning. In this work, we propose WeNLEX, a weakly supervised model for the generation of natural language explanations

Despite the success of reinforcement learning from human feedback (RLHF) in aligning language models, current reward modeling heavily relies on experimental feedback data collected from human annotators under controlled and costly conditions. In this work, we introduce observational reward modeling -- learning reward models with observational user feedback (e.g., clicks, copies, and upvotes) -- as a scalable and cost-effective alternative. We identify two fundamental challenges in this setting:

Many works in the literature show that LLM outputs exhibit discriminatory behaviour, triggering stereotype-based inferences based on the dialect in which the inputs are written. This bias has been shown to be particularly pronounced when the same inputs are provided to LLMs in Standard American English (SAE) and African-American English (AAE). In this paper, we replicate existing analyses of dialect-sensitive stereotype generation in LLM outputs and investigate the effects of mitigation strategi

Bridging the simulation-to-reality (sim2real) gap remains challenging as labelled real-world data is scarce. Existing diffusion-based approaches rely on unstructured prompts or statistical alignment, which do not capture the structured factors that make images look real. We introduce Ontology- Guided Diffusion (OGD), a neuro-symbolic zero-shot sim2real image translation framework that represents realism as structured knowledge. OGD decomposes realism into an ontology of interpretable traits -- s

Dimensionality reduction is critical for deploying dense retrieval systems at scale, yet mainstream post-hoc methods face a fundamental trade-off: principal component analysis (PCA) preserves dominant variance but underutilizes representational capacity, while whitening enforces isotropy at the cost of amplifying noise in the heavy-tailed eigenspectrum of retrieval embeddings. Intermediate spectral scaling methods unify these extremes by reweighting dimensions with a power coefficient $γ$, but t

While large language models excel in diverse domains, their performance on complex longhorizon agentic decision-making tasks remains limited. Most existing methods concentrate on designing effective reward models (RMs) to advance performance via multi-turn reinforcement learning. However, they suffer from delayed propagation in sparse outcome rewards and unreliable credit assignment with potentially overly fine-grained and unfocused turnlevel process rewards. In this paper, we propose (HISR) exp

Geometric reasoning inherently requires "thinking with constructions" -- the dynamic manipulation of visual aids to bridge the gap between problem conditions and solutions. However, existing Multimodal Large Language Models (MLLMs) are largely confined to passive inference with static diagrams, lacking the strategic knowledge of when and how to construct effective visual aids. To address this, we present a framework for Visual-Text Interleaved Chain-of-Thought. We first introduce GeoAux-Bench, t

The rapid evolution of artificial intelligence, from task-specific systems to foundation models exhibiting broad, flexible competence across reasoning, creative synthesis, and social interaction, has outpaced the conceptual and governance frameworks designed to manage it. Current regulatory paradigms, anchored in a tool-centric worldview, address algorithmic bias and transparency but leave unanswered foundational questions about what increasingly capable synthetic minds are, how societies should

Software systems have traditionally been designed for human interaction, emphasizing graphical user interfaces, usability, and cognitive alignment with end users. However, recent advances in large language model (LLM)-based agents are changing the primary consumers of software systems. Increasingly, software is no longer only used by humans, but also invoked autonomously by AI agents through structured interfaces. In this paper, we argue that software engineering is undergoing a paradigm shift f

Existing Retrieval-Augmented Generation (RAG) methods for code struggle to capture the high-level architectural patterns and cross-file dependencies inherent in complex, theory-driven codebases, such as those in algorithmic game theory (AGT), leading to a persistent semantic and structural gap between abstract concepts and executable implementations. To address this challenge, we propose Hierarchical Code/Architecture-guided Agent Generation (HCAG), a framework that reformulates repository-level

Text-guided image editors can now manipulate authentic medical scans with high fidelity, enabling lesion implantation/removal that threatens clinical trust and safety. Existing defenses are inadequate for healthcare. Medical detectors are largely black-box, while MLLM-based explainers are typically post-hoc, lack medical expertise, and may hallucinate evidence on ambiguous cases. We present MedForge, a data-and-method solution for pre-hoc, evidence-grounded medical forgery detection. We introduc

Long-form video question answering requires reasoning over extended temporal contexts, making frame selection a critical bottleneck for multi-modal large language models (MLLMs) bound by finite context windows. Within the controlled frame-budget regime that governs practical deployment, prior selectors score frames against a single global query embedding; as a result, compositional multimodal questions that involve temporal ordering or cross-modal cues such as ``what happens on screen right afte

Medical vision--language models (MVLMs) are increasingly used as perceptual backbones in radiology pipelines and as the visual front end of multimodal assistants, yet their reliability under real clinical workflows remains underexplored. Prior robustness evaluations often assume clean, curated inputs or study isolated corruptions, overlooking routine acquisition, reconstruction, display, and delivery operations that preserve clinical readability while shifting image statistics. To address this g

Text-to-image models produce images that align well with natural language prompts, but compositional generation has long been a central challenge. Models often struggle to satisfy multiple concepts within a single prompt, frequently omitting some concepts and resulting in partial success. Such failures highlight the difficulty of jointly optimizing multiple concepts during reward optimization, where competing concepts can interfere with one another. To address this limitation, we propose Correla

Persona prompting can steer LLM generation towards a domain-specific tone and pattern. This behavior enables use cases in multi-agent systems where diverse interactions are crucial and human-centered tasks require high-level human alignment. Prior works provide mixed opinions on their utility: some report performance gains when using expert personas for certain domains and their contribution to data diversity in synthetic data creation, while others find near-zero or negative impact on general u

Streaming 3D reconstruction maintains a persistent latent state that is updated online from incoming frames, enabling constant-memory inference. A key failure mode is the state update rule: aggressive overwrites forget useful history, while conservative updates fail to track new evidence, and both behaviors become unstable beyond the training horizon. To address this challenge, we propose FILT3R, a training-free latent filtering layer that casts recurrent state updates as stochastic state estima

Federated learning (FL) is severely challenged by non-independent and identically distributed (non-IID) client data, a problem that degrades global model performance, especially in multimodal perception settings. Conventional methods often fail to address the underlying semantic discrepancies between clients, leading to suboptimal performance for multimedia systems requiring robust perception. To overcome this, we introduce SemanticFL, a novel framework that leverages the rich semantic represent

Post-training alignment has produced dozens of competing algorithms -- DPO, SimPO, KTO, GRPO, and others -- yet practitioners lack controlled comparisons to guide algorithm selection. We present OXRL, a unified framework implementing 51 post-training algorithms with identical infrastructure, enabling the first large-scale apples-to-apples evaluation. Our study spans 8 algorithms across 4 model scales (0.5B--7B), 3 evaluation domains, and a 20-variant DPO taxonomy (100 runs at 1.5B, 5 seeds each)

In the era of large-scale pre-trained models, effectively adapting general knowledge to specific affective computing tasks remains a challenge, particularly regarding computational efficiency and multimodal heterogeneity. While Transformer-based methods have excelled at modeling inter-modal dependencies, their quadratic computational complexity limits their use with long-sequence data. Mamba-based models have emerged as a computationally efficient alternative; however, their inherent sequential

Collecting and annotating datasets for pixel-level semantic segmentation tasks are highly labor-intensive. Data augmentation provides a viable solution by enhancing model generalization without additional real-world data collection. Traditional augmentation techniques, such as translation, scaling, and color transformations, create geometric variations but fail to generate new structures. While generative models have been employed to extend semantic information of datasets, they often struggle t

Large vision-language models (LVLMs) demonstrate strong performance in dermatology; however, evaluating diagnostic reasoning for rare conditions remains largely unexplored. Existing benchmarks focus on common diseases and assess only final accuracy, overlooking the clinical reasoning process, which is critical for complex cases. We address this gap by constructing DermCase, a long-context benchmark derived from peer-reviewed case reports. Our dataset contains 26,030 multi-modal image-text pairs

At a time when the phenomenon of 'AI washing' is quietly spreading, an increasing number of enterprises are using the label of artificial intelligence merely as a cosmetic embellishment in their annual reports, rather than as a genuine engine driving transformation. A test regarding the essence of innovation and the authenticity of information disclosure has arrived. This paper employs large language models to conduct semantic analysis on the text of annual reports from Chinese A-share listed co

Large language models (LLMs) exhibit strong reasoning capabilities but typically require expensive post-training to reach high performance. Recent test-time alignment methods offer a lightweight alternative, but have been explored mainly for preference alignment rather than reasoning. To bridge this gap, we propose, Token-level Adaptive Routing (TARo), which steers frozen LLMs toward structured reasoning entirely at inference time. Specifically, we first train reward models on step-wise mathemat

Long-range electrostatic and polarization interactions play a central role in molecular and condensed-phase systems, yet remain fundamentally incompatible with locality-based machine-learning interatomic potentials. Although modern SO(3)-equivariant neural potentials achieve high accuracy for short-range chemistry, they cannot represent the anisotropic, slowly decaying multipolar correlations governing realistic materials, while existing long-range extensions either break SO(3) equivariance or f

Multi-agent AI systems exhibit emergent risks that no single agent produces in isolation. Existing safety frameworks rely on binary classifications of agent behavior, discarding the uncertainty inherent in proxy-based evaluation. We introduce SWARM (\textbf{S}ystem-\textbf{W}ide \textbf{A}ssessment of \textbf{R}isk in \textbf{M}ulti-agent systems), a simulation framework that replaces binary good/bad labels with \emph{soft probabilistic labels} $p = P(v{=}+1) \in [0,1]$, enabling continuous-valu

When VLMs answer correctly, do they genuinely rely on visual information? We introduce a Tri-Layer Diagnostic Framework with three per-sample metrics: Latent Anomaly Detection, Visual Necessity Score, and Competition Score, which disentangle perception, dependency, and alignment failures. Across 9 VLMs and 9,000 model-sample pairs under counterfactual blind, noise, and conflict interventions, 72.9% of samples exhibit Visual Sycophancy, a Split Beliefs pattern in which internal evidence is preser

Unsupervised Reinforcement Learning from Internal Feedback (RLIF) has emerged as a promising paradigm for eliciting the latent capabilities of Large Language Models (LLMs) without external supervision. However, current methods rely on heuristic intrinsic rewards, which often lack a well-defined theoretical optimization target and are prone to degenerative biases. In this work, we introduce PowerFlow, a principled framework that reformulates unsupervised fine-tuning as a distribution matching pro