AI SignalDemi Moore calls fighting AI a losing battle ahead of Cannes opening - Reuters
Demi Moore calls fighting AI a losing battle ahead of Cannes opening Reuters
News
Product launches, benchmarks, regulations, and the stories shaping the AI industry.
‘Shark Tank’ Star Kevin O’Leary Predicts Donald Trump Will Be Remembered as ‘The AI President’ - Yahoo
‘Shark Tank’ Star Kevin O’Leary Predicts Donald Trump Will Be Remembered as ‘The AI President’ Yahoo
OpenAI Launches Daybreak for AI-Powered Vulnerability Detection and Patch Validation - The Hacker News
OpenAI Launches Daybreak for AI-Powered Vulnerability Detection and Patch Validation The Hacker News
SAG-AFTRA Deal Stirs Concerns on Artificial Intelligence and Pensions - Variety
SAG-AFTRA Deal Stirs Concerns on Artificial Intelligence and Pensions Variety
Most Americans say AI development is moving too fast and twice as many are AI pessimists as AI optimists - YouGov
Most Americans say AI development is moving too fast and twice as many are AI pessimists as AI optimists YouGov
Google races to put Gemini at the center of Android before Apple’s AI reboot - CNBC
Google races to put Gemini at the center of Android before Apple’s AI reboot CNBC
IIM Indore to roll out new AI, machine learning courses from 2026-27 - The Times of India
IIM Indore to roll out new AI, machine learning courses from 2026-27 The Times of India
Florida students boo graduation speaker who called AI ‘next Industrial Revolution’ - The Guardian
Florida students boo graduation speaker who called AI ‘next Industrial Revolution’ The Guardian
Elon Musk said control of OpenAI should go to his children, Sam Altman tells jury - BBC
Elon Musk said control of OpenAI should go to his children, Sam Altman tells jury BBC
Belief or Circuitry? Causal Evidence for In-Context Graph Learning
arXiv:2605.08405v1 Announce Type: new Abstract: How do LLMs learn in-context? Is it by pattern-matching recent tokens, or by inferring latent structure? We probe this question using a toy graph random-walk across two competing graph structures. This task's answer is, in principle, decidable: either the model tracks global topology, or it copies local transitions. We present two lines of evidence that neither account alone is sufficient. First, reconstructing the internal representation structure via PCA reveals that at intermediate mixture ratios, both graph topologies are encoded in orthogonal principal subspaces simultaneously. This pattern is difficult to reconcile with purely local transition copying. Second, residual-stream activation patching and graph-difference steering causally intervene on this graph-family signal: late-layer patching almost fully transfers the clean graph preference, while linear steering moves predictions in the intended direction and fails under norm-matched and label-shuffled controls. Taken together, our findings are most consistent with a dual-mechanism account in which genuine structure inference and induction circuits operate in parallel.
CoCoDA: Co-evolving Compositional DAG for Tool-Augmented Agents
arXiv:2605.08399v1 Announce Type: new Abstract: Tool-augmented language models can extend small language models with external executable skills, but scaling the tool library creates a coupled challenge: the library must evolve with the planner as new reusable subroutines emerge, while retrieval from the growing library must remain within a fixed context budget. Existing tool-use and skill-library methods typically treat tools as flat or text-indexed memories, causing prompt cost to grow with library size and obscuring the typed, compositional structure of executable code. We propose CoCoDA, a framework that co-evolves the planner and tool library through a single code-native structure: a compositional code DAG. Nodes are primitive or composite tools, edges encode invocation dependencies, and each node stores a typed signature, description, pre/post-condition specification, and worked examples. At inference time, Typed DAG Retrieval prunes candidates by symbolic signature unification, ranks survivors by descriptions, filters them by behavioral specifications, and disambiguates with examples, keeping expensive context materialization on progressively smaller candidate sets. At training time, successful trajectories are folded into validated composite tools, while the planner is updated with a DAG-induced reward that credits composites by their primitive expansion size. We provide theoretical results showing retrieval cost reduction, sublinear retrieval time, compositional advantage under the shaped reward, monotone co-evolution under conservative updates, and DAG well-formedness. Across mathematical reasoning, tabular analysis, and code task benchmarks, CoCoDA enables an 8B student to match or exceed a 32B teacher on GSM8K and MATH and consistently improves over strong tool-use and library-learning baselines.
PLACO: A Multi-Stage Framework for Cost-Effective Performance in Human-AI Teams
arXiv:2605.08388v1 Announce Type: new Abstract: Human-AI teams play a pivotal role in improving overall system performance when neither the human nor the model can achieve such performance on their own. With the advent of powerful and accessible Generative AI models, several mundane tasks have morphed into Human-AI team tasks. From writing essays to developing advanced algorithms, humans have found that using AI assistance has led to an accelerated work pace like never before. In classification tasks, where the final output is a single hard label, it is crucial to address the combination of human and model output. Prior work elegantly solves this problem using Bayes rule, using the assumption that human and model output are conditionally independent given the ground truth. Specifically, it discusses a combination method to combine a single deterministic labeler (the human) and a probabilistic labeler (the classifier model) using the model's instance-level and the human's class-level calibrated probabilities.
SkillLens: Adaptive Multi-Granularity Skill Reuse for Cost-Efficient LLM Agents
arXiv:2605.08386v1 Announce Type: new Abstract: Skill libraries have become a practical way for LLM agents to reuse procedural experience across tasks. However, existing systems typically treat skills as flat, single-resolution prompt blocks. This creates a tension between relevance and cost: injecting coarse skills can introduce irrelevant or misleading context, while rewriting entire skills is expensive and often unnecessary. We propose SkillLens, a hierarchical skill-evolution framework that organizes skills into a four-layer graph of policies, strategies, procedures, and primitives, and retrieves them at mixed granularity. Given a task, SkillLens first retrieves semantically relevant skill seeds, expands them through degree-corrected random walk over the skill graph, and then uses a verifier to decide whether each visited unit should be accepted, decomposed, rewritten, or skipped. This enables the agent to reuse compatible subskills directly while adapting only locally mismatched components. To improve the system over time, SkillLens further refines multi-granularity skills and verifier in order to improve its routing decisions. We provide theoretical analysis showing that mixed-granularity adaptation incurs sublinear cost under sparse mismatch assumptions and that the evolutionary update rule monotonically improves the validation objective until a local optimum. Across MuLocbench and ALFWorld, SkillLens consistently improves over strong skill-based baselines, achieving up to a 6.31 percentage-point Acc@1 gain for bug localization and raising agent success rate from 45.00% to 51.31%.
MemQ: Integrating Q-Learning into Self-Evolving Memory Agents over Provenance DAGs
arXiv:2605.08374v1 Announce Type: new Abstract: Episodic memory allows LLM agents to accumulate and retrieve experience, but current methods treat each memory independently, i.e., evaluating retrieval quality in isolation without accounting for the dependency chains through which memories enable the creation of future memories. We introduce MemQ, which applies TD($\lambda$) eligibility traces to memory Q-values, propagating credit backward through a provenance DAG that records which memories were retrieved when each new memory was created. Credit weight decays as $(\gamma\lambda)^d$ with DAG depth $d$, replacing temporal distance with structural proximity. We formalize the setting as an Exogenous-Context MDP, whose factored transition decouples the exogenous task stream from the endogenous memory store. Across six benchmarks, spanning OS interaction, function calling, code generation, multimodal reasoning, embodied reasoning, and expert-level QA, MemQ achieves the highest success rate on all six in generalization evaluation and runtime learning, with gains largest on multi-step tasks that produce deep and relevant provenance chains (up to +5.7~pp) and smallest on single-step classification (+0.77~pp) where single-step updates already suffice. We further study how $\gamma$ and $\lambda$ interact with the EC-MDP structure, providing principled guidance for parameter selection and future research. Code will be available soon.
On Distinguishing Capability Elicitation from Capability Creation in Post-Training: A Free-Energy Perspective
arXiv:2605.08368v1 Announce Type: new Abstract: Debates about large language model post-training often treat supervised fine-tuning (SFT) as imitation and reinforcement learning (RL) as discovery. But this distinction is too coarse. What matters is whether a training procedure increases the probability of behaviors the pretrained model could already produce, or whether it changes what the model can practically reach. We argue that post-training research should distinguish between capability elicitation and capability creation. We make this distinction operational by introducing the notion of accessible support: the set of behaviors that a model can practically produce under finite budgets. Post-training that reweights behaviors within this support is capability elicitation; whereas changing the support itself corresponds to capability creation. We develop this argument through a free-energy view of post-training. SFT and RL can both be seen as reweighting a pretrained reference distribution, only with different external signals. Demonstration signals define low-energy behavior for SFT, and reward signals define low-energy behavior for RL. When the update remains close to the base model, the main effect is local reweighting, not capability creation. Within this framework, the central question is no longer whether post-training is framed as SFT or RL, but whether it reweights behaviors already within reach, or instead expands the model's reachable behavioral space through search, interaction, tool use, or the incorporation of new information.
Embeddings for Preferences, Not Semantics
arXiv:2605.08360v1 Announce Type: new Abstract: Modern AI is opening the door to collective decision-making in which participants express their views as free-form text rather than voting on a fixed set of candidates. A natural idea is to embed these opinions in a vector space so that the substantial literature on facility location problems and fair clustering can be brought to bear. But standard text embeddings measure semantic similarity, whereas distances in facility location problems and fair clustering require what we call \textit{preferential similarity}: a participant's agreement with a piece of text should be inversely related to their distance from it. Off-the-shelf embeddings inherit a coarse preference signal through a correlation between semantic and preferential similarity, but fail to capture preferences when the correlation breaks. We formalize this as an invariance problem: text embedding models encode both a preference-relevant signal (stance and values) and semantic nuisance (style and wording), and the two are observationally correlated, so a geometry that relies on nuisance can appear preference-correct even when it is not. We show that synthetic training data designed to break this correlation provably shifts the optimal scorer away from nuisance-dominated cosine and significantly improves preference prediction across 11 online deliberation datasets.
Auto-Rubric as Reward: From Implicit Preferences to Explicit Multimodal Generative Criteria
arXiv:2605.08354v1 Announce Type: new Abstract: Aligning multimodal generative models with human preferences demands reward signals that respect the compositional, multi-dimensional structure of human judgment. Prevailing RLHF approaches reduce this structure to scalar or pairwise labels, collapsing nuanced preferences into opaque parametric proxies and exposing vulnerabilities to reward hacking. While recent Rubrics-as-Reward (RaR) methods attempt to recover this structure through explicit criteria, generating rubrics that are simultaneously reliable, scalable, and data-efficient remains an open problem. We introduce Auto-Rubric as Reward (ARR), a framework that reframes reward modeling from implicit weight optimization to explicit, criteria-based decomposition. Before any pairwise comparison, ARR externalizes a VLM's internalized preference knowledge as prompt-specific rubrics, translating holistic intent into independently verifiable quality dimensions. This conversion of implicit preference structure into inspectable, interpretable constraints substantially suppresses evaluation biases including positional bias, enabling both zero-shot deployment and few-shot conditioning on minimal supervision. To extend these gains into generative training, we propose Rubric Policy Optimization (RPO), which distills ARR's structured multi-dimensional evaluation into a robust binary reward, replacing opaque scalar regression with rubric-conditioned preference decisions that stabilize policy gradients. On text-to-image generation and image editing benchmarks, ARR-RPO outperforms pairwise reward models and VLM judges, demonstrating that explicitly externalizing implicit preference knowledge into structured rubrics achieves more reliable, data-efficient multimodal alignment, revealing that the bottleneck is the absence of a factorized interface, not a deficit of knowledge.
Spatial Priming Outperforms Semantic Prompting: A Grid-Based Approach to Improving LLM Accuracy on Chart Data Extraction
arXiv:2605.08220v1 Announce Type: new Abstract: The automated extraction of data from scientific charts is a critical task for large-scale literature analysis. While multimodal Large Language Models (LLMs) show promise, their accuracy on non-standardized charts remains a challenge. This raises a key research question: what is the most effective strategy to improve model performance (high-level semantic priming) or low-level spatial priming? This paper presents a comparative investigation into these two distinct strategies. We describe our exploratory experiments with semantic methods, such as a two-stage metadata-first framework and Chain-of-Thought, which failed to produce a statistically significant improvement. In contrast, we present a simple but highly effective spatial priming method: overlaying a coordinate grid onto the chart image before analysis. Our quantitative experiment on a synthetic dataset demonstrates that this grid-based approach provides a statistically significant reduction in data extraction error (SMAPE reduced from 25.5% to 19.5%, p < 0.05) compared to a baseline. We conclude that for the current generation of multimodal models, providing explicit spatial context is a more effective and reliable strategy than high-level semantic guidance for this class of tasks.
Where Reliability Lives in Vision-Language Models: A Mechanistic Study of Attention, Hidden States, and Causal Circuits
arXiv:2605.08200v1 Announce Type: new Abstract: A pervasive intuition holds that vision-language models (VLMs) are most trustworthy when their attention maps look sharp: concentrated attention on the queried region should imply a confident, calibrated answer. We test this Attention-Confidence Assumption directly. We instrument three open-weight VLM families (LLaVA-1.5, PaliGemma, Qwen2-VL; 3-7B parameters) with a unified mechanistic pipeline -- the VLM Reliability Probe (VRP) -- that compares attention structure, generation dynamics, and hidden-state geometry against a single correctness label. Three results emerge. (i) Attention structure is a near-zero predictor of correctness (R_pb(C_k,y)=0.001, 95% CI [-0.034,0.036]; R_pb(H_s,y)=-0.012, [-0.047,0.024] on a pooled n=3,090 split), even though attention remains causally necessary for feature extraction (top-30% patch masking drops accuracy by 8.2-11.3 pp, p 0.95 on POPE for two of three families, and self-consistency at K=10 is the strongest behavioral predictor we measure at 10x inference cost (R_pb=0.43). (iii) Causal neuron-level ablations expose a sharp architectural split with direct monitor-design implications: late-fusion LLaVA concentrates reliability in a fragile late bottleneck (-8.3 pp object-identification accuracy after top-5 probe-neuron ablation), whereas early-fusion PaliGemma and Qwen2-VL distribute it widely and absorb destruction of ~50% of their peak-layer hidden dimension with <=1 pp degradation. The takeaway is narrow but consequential: in 3-7B VLMs, reliability is read more reliably off hidden-state geometry, layer-wise margin formation, and sparse late-layer circuits than off attention-map sharpness.
REVA Academy for Corporate Excellence Strengthens Industry-Ready Talent with Master’s Programs in Artificial Intelligence and Business Analytics - ThePrint
REVA Academy for Corporate Excellence Strengthens Industry-Ready Talent with Master’s Programs in Artificial Intelligence and Business Analytics ThePrint
Hackers pushing innovation in AI-enabled hacking operations, Google says - ET Telecom
Hackers pushing innovation in AI-enabled hacking operations, Google says ET Telecom
40% of firms deploying Artificial Intelligence to use AI observability tools - Deccan Herald
40% of firms deploying Artificial Intelligence to use AI observability tools Deccan Herald
China Sought Access to Anthropic’s Newest A.I. The Answer Was No. - The New York Times
China Sought Access to Anthropic’s Newest A.I. The Answer Was No. The New York Times
Google says hackers used AI to develop a major security flaw - Politico
Google says hackers used AI to develop a major security flaw Politico
Trading Day: AI beats Iran impasse - Reuters
Trading Day: AI beats Iran impasse Reuters
Some AI tools are getting dumber and more expensive
In recent weeks, AI companies have either tightened up or hiked prices on their subscription services—or throttled access to the systems they offer.
Introducing Claude Platform on AWS: Anthropic’s native platform, through your AWS account - Amazon Web Services (AWS)
Introducing Claude Platform on AWS: Anthropic’s native platform, through your AWS account Amazon Web Services (AWS)
This AI Lamp Scam Needs To Stop.
This AI Lamp Scam Needs To Stop.
ChatGPT can’t decide which days have D…
ChatGPT can’t decide which days have D…
$300k PC is CRAZY #pc #datacenter #pcs #nvidia
$300k PC is CRAZY #pc #datacenter #pcs #nvidia
Google Says Criminal Hackers Used A.I. to Find a Major Software Flaw - The New York Times
Google Says Criminal Hackers Used A.I. to Find a Major Software Flaw The New York Times
Strategies of high-accuracy memristor-based analogue computing in memory for artificial intelligence - Nature
Strategies of high-accuracy memristor-based analogue computing in memory for artificial intelligence Nature
Tech stocks today: Chipmaker Cerebras to stage blockbuster IPO, AI in focus for Trump-Xi meeting - Yahoo Finance
Tech stocks today: Chipmaker Cerebras to stage blockbuster IPO, AI in focus for Trump-Xi meeting Yahoo Finance
Analysis | See the hidden rules behind AI. Then use them to rewrite this article. - The Washington Post
Analysis | See the hidden rules behind AI. Then use them to rewrite this article. The Washington Post
Forget the AI job apocalypse. AI’s real threat is worker control and surveillance - The Guardian
Forget the AI job apocalypse. AI’s real threat is worker control and surveillance The Guardian
Hackers Used AI to Develop First Known Zero-Day 2FA Bypass for Mass Exploitation - The Hacker News
Hackers Used AI to Develop First Known Zero-Day 2FA Bypass for Mass Exploitation The Hacker News
AI-powered hacking has exploded into industrial-scale threat, Google says - The Guardian
AI-powered hacking has exploded into industrial-scale threat, Google says The Guardian
Google says it likely thwarted effort by hacker group to use AI for 'mass exploitation event' - CNBC
Google says it likely thwarted effort by hacker group to use AI for 'mass exploitation event' CNBC
When Does Critique Improve AI-Assisted Theoretical Physics? SCALAR: Structured Critic--Actor Loop for Agentic Reasoning
arXiv:2605.06772v1 Announce Type: new Abstract: As large language models (LLMs) show increasing promise on research-level physics reasoning tasks and agentic AI becomes more common, a practical question emerges: How does the interaction between researchers and agents affect the results? We study this using SCALAR (Structured Critic--Actor Loop for AI Reasoning), an Actor--Critic--Judge pipeline applied to quantum field theory and string theory problems. The Actor proposes solutions, the Critic provides iterative feedback, and an independent Judge evaluates the transcript against reference solutions. We vary the Actor persona, the Critic feedback strategy, and the Actor model family and scale. Multi-turn dialogue improves over single-shot attempts throughout, but both the mechanism of improvement and the value of different prompting choices depend strongly on the Actor--Critic pairing. Increasing the scale within one model family (e.g. from the 8B-parameter DeepSeek-R1 variant to DeepSeek-R1 70B) improves some easier-problem behavior, but does not remove the hardest bottleneck we observe. Critic feedback strategy matters most clearly in the asymmetric Actor--Critic setting (e.g., a lightweight Haiku Actor guided by a stronger Sonnet Critic), where constructive feedback improves mean-score outcomes. In same-family Actor--Critic settings, strategy effects are weaker: lenient feedback is sometimes favored, while strict and adversarial feedback are not beneficial. Taken together, SCALAR provides a controlled testbed for evaluating which interaction structures help or hinder AI-driven scientific discovery.
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
arXiv:2605.06761v1 Announce Type: new Abstract: The web is complex, open-ended, and constantly changing, making it challenging to scale training data for visual web agents. Existing data collection attempts remain limited to offline trajectories for supervised fine-tuning or a handful of simulated environments for RL training, thus failing to capture web diversity. We propose Weblica (Web Replica), a framework for constructing reproducible and scalable web environments. Our framework leverages 1) HTTP-level caching to capture and replay stable visual states while preserving interactive behavior and 2) LLM-based environment synthesis grounded in real-world websites and core web navigation skills. Using this framework, we scale RL training to thousands of diverse environments and tasks. Our best model, Weblica-8B, outperforms open-weight baselines of similar size across multiple web navigation benchmarks while using fewer inference steps, scales favorably with additional test-time compute, and is competitive with API models.
From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms
arXiv:2605.06716v1 Announce Type: new Abstract: Large Language Model (LLM)-based agents have fundamentally reshaped artificial intelligence by integrating external tools and planning capabilities. While memory mechanisms have emerged as the architectural cornerstone of these systems, current research remains fragmented, oscillating between operating system engineering and cognitive science. This theoretical divide prevents a unified view of technological synthesis and a coherent evolutionary perspective. To bridge this gap, this survey proposes a novel evolutionary framework for LLM agent memory mechanisms, formalizing the development process into three stages: Storage (trajectory preservation), Reflection (trajectory refinement), and Experience (trajectory abstraction). We first formally define these three stages before analyzing the three core drivers of this evolution: the necessity for long-range consistency, the challenges in dynamic environments, and the ultimate goal of continual learning. Furthermore, we specifically explore two transformative mechanisms in the frontier Experience stage: proactive exploration and cross-trajectory abstraction. By synthesizing these disparate views, this work offers robust design principles and a clear roadmap for the development of next-generation LLM agents.
CASCADE: Case-Based Continual Adaptation for Large Language Models During Deployment
arXiv:2605.06702v1 Announce Type: new Abstract: Large language models (LLMs) have become a central foundation of modern artificial intelligence, yet their lifecycle remains constrained by a rigid separation between training and deployment, after which learning effectively ceases. This limitation contrasts with natural intelligence, which continually adapts through interaction with its environment. In this paper, we formalise deployment-time learning (DTL) as the third stage in the LLM lifecycle that enables LLM agents to improve from experience during deployment without modifying model parameters. We present CASCADE (CASe-based Continual Adaptation during DEployment), a general and principled framework that equips LLM agents with an explicit, evolving episodic memory. CASCADE formulates experience reuse as a contextual bandit problem, enabling principled exploration-exploitation trade-offs and establishing no-regret guarantees over long-term interactions. This design allows agents to accumulate, select, and refine task-relevant cases, transforming past experience into actionable knowledge. Across 16 diverse tasks spanning medical diagnosis, legal analysis, code generation, web search, tool use, and embodied interaction, CASCADE improves macro-averaged success rate by 20.9% over zero-shot prompting while consistently outperforming gradient-based and memory-based baselines. By reframing deployment as an adaptive learning process, this work establishes a foundation for continually improving AI systems.
Hidden Coalitions in Multi-Agent AI: A Spectral Diagnostic from Internal Representations
arXiv:2605.06696v1 Announce Type: new Abstract: Collections of interacting AI agents can form coalitions, creating emergent group-level organization that is critical for AI safety and alignment. However, observing agent behavior alone is often insufficient to distinguish genuine informational coupling from spurious similarity, as consequential coalitions may form at the level of internal representations before any overt behavioral change is apparent. Here, we introduce a practical method for detecting coalition structure from the internal neural representations of multi-agent systems. The approach constructs a pairwise mutual-information graph from the hidden states of agents and applies spectral partitioning to identify the most salient coalition boundary. We validate this method in two domains. First, in multi-agent reinforcement learning environments, the method successfully recovers programmed hierarchical and dynamic coalition structures and correctly rejects false positives arising from behavioral coordination without informational coupling. Second, using a large language model, the method identifies coalition structures implied by descriptive prompts, tracks dynamic team reassignments, and reveals a representational hierarchy where explicit labels dominate over conflicting interaction patterns. Across both settings, the recovered partition reveals subgroup organization that a scalar cross-agent mutual-information measure cannot distinguish. The results demonstrate that analyzing hidden-state mutual information through spectral partitioning provides a scalable diagnostic for identifying representational coalitions, offering a valuable tool for monitoring emergent structure in distributed AI systems.
State Representation and Termination for Recursive Reasoning Systems
arXiv:2605.06690v1 Announce Type: new Abstract: Recursive reasoning systems alternate between acquiring new evidence and refining an accumulated understanding. Two design choices are typically left implicit: how to represent the evolving reasoning state, and when to stop iterating. This paper addresses both. We represent the reasoning state as an epistemic state graph encoding extracted claims, evidential relations, open questions, and confidence weights. We define the order-gap as the distance between the states reached by expand-then-consolidate versus consolidate-then-expand; a small order-gap suggests that the two orderings agree and further iteration is unlikely to help. Our main result gives a necessary and sufficient condition for the linearised order-gap to be non-degenerate near the fixed point, showing when the criterion is informative rather than algebraically vacuous. This is a local condition, not a global convergence guarantee. We apply the framework to recursive reasoning systems and sketch its application to agent loops, tree-of-thought reasoning, theorem proving, and continual learning.
Fast and Effective Redistricting Optimization via Composite-Move Tabu Search
arXiv:2605.06682v1 Announce Type: new Abstract: Spatial redistricting is a practical combinatorial optimization problem that demands high-quality solutions, rapid turnaround, and flexibility to accommodate multi-criteria objectives and interactive refinement. A central challenge is the contiguity constraint: enforcing contiguity in integer-programming or heuristic search can severely shrink the feasible neighborhood, weaken exploration, and trap the search in poor local optima. We introduce a composite-move Tabu search (CM-Tabu) that systematically expands the feasible neighborhood space in Tabu search while preserving contiguity. When a boundary unit cannot be reassigned individually without disconnecting its district, our method identifies a minimal set of units that can move together, or a pair of units (or sets of units) that can be switched, as a contiguity-preserving composite move. Candidate single-unit and composite moves are generated in linear time by analyzing each district's contiguity graph using articulation points and biconnected components. Extensive experiments demonstrate that the proposed approach substantially improves solution quality, run-to-run robustness, and computational efficiency relative to traditional Tabu search and other baselines. For example, in the Philadelphia case, the approach can consistently attain the theoretical global optimum in population-equality and support multi-criteria trade-offs. CM-Tabu delivers optimization performance suitable for real-world practices and decision-support workflows.
When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization Commitment
arXiv:2605.06723v1 Announce Type: new Abstract: Language models often generate reasoning before giving a final answer, but the visible answer does not reveal when the model's answer preference became stable. We study this question through a narrow computable object: \emph{finite-answer preference stabilization}. For a model state and specified answer verbalizers, we project the model's own continuation probabilities onto a finite answer set; in binary tasks this yields an exact log-odds code, $\delta(\xi)=S_\theta(\mathrm{yes}\mid\xi)-S_\theta(\mathrm{no}\mid\xi)$. This target defines parser-based answer onset, retrospective stabilization time, and lead without relying on greedy rollouts or learned probes. In controlled delayed-verdict tasks with Qwen3-4B-Instruct, the contextual finite-answer projection stabilizes before the answer is parseable, with 17--31 token mean lead in the main templates and positive, shorter lead in a parser-clean replication. The signal tracks the model's eventual output rather than truth, is linearly recoverable from compact hidden summaries, is partly separable from cursor progress, and transfers as shared information without a single invariant coordinate. Diagnostics separate the measurement from online stopping, verbalizer-free belief, and causal answer control; exact steering shows local sensitivity of $\delta$ but not reliable generation control.
More Thinking, More Bias: Length-Driven Position Bias in Reasoning Models
arXiv:2605.06672v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning and reasoning-tuned models such as DeepSeek-R1 are commonly assumed to reduce shallow heuristic biases by thinking carefully. We test this on position bias in multiple-choice QA and find a different story: within any reasoning-capable model, per-question position bias scales with the length of the reasoning trajectory. Across thirteen reasoning-mode configurations (two R1-distilled 7-8B models, two base models prompted with CoT, and DeepSeek-R1 at 671B) on MMLU, ARC-Challenge, and GPQA, twelve show a positive partial correlation between trajectory length and Position Bias Score (PBS) after controlling for accuracy, ranging from 0.11 to 0.41 (all p < 0.05). All twelve open-weight reasoning-mode configurations show monotonically increasing PBS across length quartiles. A truncation intervention provides causal evidence: continuations resumed from later points in the trajectory are increasingly likely to shift toward position-preferred options (16% to 32% for R1-Qwen-7B across absolute-position buckets). At 671B, aggregate PBS collapses to 0.019, but the length effect still manifests in the longest quartile (PBS = 0.071), suggesting that accuracy gates the expression of length-driven bias rather than eliminating the underlying mechanism. We additionally find that direct-answer position bias is a distinct phenomenon with a different footprint (strong in Llama-Instruct-direct, weak in Qwen-Instruct-direct, and uncorrelated with trajectory length): CoT reasoning replaces this baseline bias with length-accumulated bias. Our results argue that reasoning-capable models should not be treated as order-robust by default in MCQ evaluation pipelines, and offer a diagnostic toolkit (PBS, commitment change point, effective switching, truncation probes) for auditing position bias in reasoning models.
GraphDC: A Divide-and-Conquer Multi-Agent System for Scalable Graph Algorithm Reasoning
arXiv:2605.06671v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated strong potential for many mathematical problems. However, their performance on graph algorithmic tasks is still unsatisfying, since graphs are naturally more complex in topology and often require systematic multi-step reasoning, especially on larger graphs. Motivated by this gap, we propose GraphDC, a Divide-and-Conquer multi-agent framework for scalable graph algorithm reasoning. Specifically, inspired by Divide-and-Conquer design, GraphDC decomposes an input graph into smaller subgraphs, assigns each subgraph to a specialized agent for local reasoning, and uses a master agent to integrate the local outputs with inter-subgraph information to produce the final solution. This hierarchical design reduces the reasoning burden on individual agents, alleviates computational bottlenecks, and improves robustness on large graph instances. Extensive experiments show that GraphDC consistently outperforms existing methods on graph algorithm reasoning across diverse tasks and scales, especially on larger instances where direct end-to-end reasoning is less reliable.
Tech is turning increasingly to religion in a quest to create ethical AI - The Indian Express
Tech is turning increasingly to religion in a quest to create ethical AI The Indian Express
OpenAI staff cashed out 6.6 billion dollars in one day, says report. How much richer did it make the employees? - Mint
OpenAI staff cashed out 6.6 billion dollars in one day, says report. How much richer did it make the employees? Mint
Mistaking AI behaviour for conscious being | Letter - The Guardian
Mistaking AI behaviour for conscious being | Letter The Guardian
Do you need a chief AI officer? Here's how the tech is changing boardrooms - CNBC
Do you need a chief AI officer? Here's how the tech is changing boardrooms CNBC
Anthropic links Claude’s blackmail behaviour to ‘evil AI’ portrayals online - Indiatimes
Anthropic links Claude’s blackmail behaviour to ‘evil AI’ portrayals online Indiatimes
The new AI-powered Google Finance is expanding to Europe.
This week, the new, AI-powered Google Finance is launching across Europe, with full local language support. This reimagined experience offers a suite of powerful capabil…
Reimagining the classroom in the age of artificial intelligence - The Tribune
Reimagining the classroom in the age of artificial intelligence The Tribune
Prediction: This Nvidia-Backed Artificial Intelligence (AI) Company Might Be the Next Meme Stock. Here's Why It's Actually a Buy at $13 Per Share. - The Motley Fool
Prediction: This Nvidia-Backed Artificial Intelligence (AI) Company Might Be the Next Meme Stock. Here's Why It's Actually a Buy at $13 Per Share. The Motley Fool
AI data centers face increasing complaints about inaudible but 'felt' infrasound — citizens complain high- and low-frequency sounds do not register on decibel meters but cause adverse health effects - Tom's Hardware
AI data centers face increasing complaints about inaudible but 'felt' infrasound — citizens complain high- and low-frequency sounds do not register on decibel meters but cause adverse health effects Tom's Hardware
Are SanDisk and Micron Too Expensive? Here's How You Can Invest in the Artificial Intelligence (AI) Memory Supercycle for Just $50. - Yahoo Finance
Are SanDisk and Micron Too Expensive? Here's How You Can Invest in the Artificial Intelligence (AI) Memory Supercycle for Just $50. Yahoo Finance
Algo rhythm: Indian firms lack basics to use AI for cybersecurity - Business Standard
Algo rhythm: Indian firms lack basics to use AI for cybersecurity Business Standard
Vibe Coded Apps Are Spilling Users' Personal Information Directly Into the Maw of Greedy Hackers - Futurism
Vibe Coded Apps Are Spilling Users' Personal Information Directly Into the Maw of Greedy Hackers Futurism
Alibaba to integrate Qwen AI with Taobao, launch agentic shopping - The Indian Express
Alibaba to integrate Qwen AI with Taobao, launch agentic shopping The Indian Express
Top AI Graduate Programs for Building a Successful Career in Artificial Intelligence - Investopedia
Top AI Graduate Programs for Building a Successful Career in Artificial Intelligence Investopedia
IIT-M centre to train one million teachers on AI by 2027 - Indiatimes
IIT-M centre to train one million teachers on AI by 2027 Indiatimes
I knew my writing students were using AI. Their confessions led to a powerful teaching moment | Micah Nathan - The Guardian
I knew my writing students were using AI. Their confessions led to a powerful teaching moment | Micah Nathan The Guardian
How to test your filament #3dprinting #3dprinted #ai3dmodelgenerator
Do these filament tests before large-scale printing: 1. Temperature test 2. Flow test 3. Speed test Of course, don’t forget to try Tripo AI and DIY your own models!
When did AirPods become normal? #carterpcs #tech #airpods #apple
When did AirPods become normal? #carterpcs #tech #airpods #apple
Opinion | The Atheist and the Machine God - The New York Times
Opinion | The Atheist and the Machine God The New York Times
So you’ve heard these AI terms and nodded along; let’s fix that - TechCrunch
So you’ve heard these AI terms and nodded along; let’s fix that TechCrunch
600% AI Surge In Just 3 Months: Why Cloudflare Is Cutting Over 1,100 Jobs - NDTV
600% AI Surge In Just 3 Months: Why Cloudflare Is Cutting Over 1,100 Jobs NDTV
Meta’s Embrace of A.I. Is Making Its Employees Miserable - The New York Times
Meta’s Embrace of A.I. Is Making Its Employees Miserable The New York Times
See what happens when creative legends use AI to make ads for small businesses.
Today we're launching The Small Brief, an initiative bringing together three ad industry icons to champion a local businesses they love. Their mission is to build breakt…
Roche shells out up to $1B for PathAI in move to boost artificial intelligence-powered diagnostics - Fierce Biotech
Roche shells out up to $1B for PathAI in move to boost artificial intelligence-powered diagnostics Fierce Biotech
5 gardening tips you can try right in Search
We’ve rounded up the top ways you can use Google’s AI Mode, Search Live and Shopping to help your plants thrive.
Frontier models weekly: efficiency, multimodal, and open weights
New releases squeezed more capability per token — here is what builders should prioritize this month.
Google is partnering with XPRIZE and Range Media Partners on the $3.5 million Future Vision film competition.
Google is partnering with XPRIZE and Range Media Partners on the $3.5 million Future Vision film competition.
Reduce friction and latency for long-running jobs with Webhooks in Gemini API
Event-Driven Webhooks are a push-based notification system that eliminates the need for inefficient polling.
The latest AI news we announced in April 2026
Here are Google’s latest AI updates from April 2026
World's smallest 4k headset 😎 Visor.com #tech #vr #technology #virtualreality #Thesupermariogalaxy
Immersed built the #1 AR/VR app with 1.5M+ users, the sleekest Visor, & an AI agent that transforms how you work.
Celebrating 20 years of Google Translate: Fun facts, tips and new features to try
Google’s sharing 20 fun facts to celebrate Google Translate turning 20, from its roots as a 2006 AI experiment to supporting almost 250 languages today.
Join the new AI Agents Vibe Coding Course from Google and Kaggle
Google is bringing back its 5-Day AI Agents Intensive Course with Kaggle and registration is open.
8 Gemini tips for organizing your space (and life)
Organize your home and digital space with Gemini. Use AI-powered tips for cleaning schedules, inbox decluttering, seasonal chores.
Elevating Austria: Google invests in its first data center in the Alps.
Google has been a proud part of Austria’s landscape for years, and today, we’re announcing our first data center in Kronstorf, generating 100 direct jobs. This facility …
Here’s how our TPUs power increasingly demanding AI workloads.
Learn how Google’s TPUs power increasingly demanding AI workloads with this new video.
News: a practical EU AI Act checklist for product teams
What to document now so you are not translating chaos into policy language later.
Tutorial on How to Curve Any Monitor
PC Inquiry🔨: https://archetypeorigins.com/inquiries Links⚫⚪: https://linktr.ee/archetype_origins Buy us a Coffee! ☕ https://www.buymeacoffee.com/ArchetypeOrigins