Claude Code v2.1.181 introduces support for setting config settings via prompt syntax like /config thinking=false, adds sandbox Apple Events support on macOS, and improves streaming, auto-retry, and subagent behavior. It also fixes numerous bugs related to startup, file handling, clipboard, and UI responsiveness across platforms.
Claude v2.1.178 introduces new permission rules using Tool(param:value) syntax, improved workflow and skill loading in nested directories, and enhanced auto mode and error messaging. It fixes critical issues including crashes, authentication errors, and UI behavior in Chrome and VSCode, while refining tool prompts and undo functionality.
PragReST is a self-supervised framework that enhances large language models' pragmatic reasoning by generating counterfactual reasoning traces and training via supervised fine-tuning and reinforcement learning. It outperforms baseline models on four pragmatic benchmarks, improving Qwen3-8B and Qwen3-14B by 5.37% and 5-5.50% accuracy respectively, and maintains strong performance on general-knowledge and mathematical reasoning tasks.
A new study introduces VETO, a benchmark of 2,032 BBQ-derived contrastive pairs, to quantify misfired alignment in large language models. It defines the Misfired Alignment Rate (MAR) and finds that all benchmarked LLMs exhibit MARs between 4.7% and 18.9%, while human participants achieve 0%. The research shows alignment cues can amplify these failures, with evidence suppression occurring in late layers of models and emerging after instruction training.
The Frustrated Synchronization Network (FSN) achieves lower validation loss than a RoPE-SwiGLU transformer at every epoch on character-level text and code tasks. At one million parameters, FSN converges to a validation loss of 1.5953 ± 0.0014, outperforming the transformer's converged loss of 1.611. This advantage persists up to four million parameters, with ongoing evaluations beyond that scale.
A new method called output vector editing minimally modifies MLP neurons' output vectors to suppress memorized sequences in large language models, achieving up to 87.9% suppression in OLMo-7B. This approach outperforms zeroing neuron activations by a factor of 2.7 and works across four models from 36-7B parameters, with success rates scaling with model size and showing consistent performance across architectures.
HandwritingAgent synthesizes natural handwriting in SVG format without style-specific training. It uses a large reasoning model to generate stroke sequences in a grid canvas, conditioned on text input and a reference style image, enabling efficient, controllable, and generalizable handwriting generation.
A data-centric approach improves long-context reasoning in large language models, using eight curated datasets with 14K examples across retrieval, multi-evidence synthesis, and reasoning tasks. When paired with minimal outcome-based GRPO training, it achieves average gains of +7.2 to +6.4 points on seven benchmarks, outperforming prior RL training sets, and enhances agentic performance by +4.8 and +7.0 points on GAIA and BrowseComp respectively.
REVES introduces a two-stage iterative framework that enhances large language model reasoning through sequential revision and verification. It achieves +6.5 points over RL baselines and +4.0 points over standard multi-turn training on LiveCodeBench, using a 4B base model with fewer rollouts than larger systems. The method improves error correction and generalizes to out-of-distribution puzzles like n_queens and mini_sudoku.
SenFlow introduces a novel method for detecting AI-generated text in hybrid documents by modeling inter-sentence dependencies. It achieves state-of-the-art performance on MOSAIC, a benchmark of 16,000 documents from PubMed and XSum, with a +4.15 pp Macro-F1 gain on cross-domain transfer. SenFlow reveals that AI-generated content still exhibits generator-dependent sentence-length patterns, exploitable by sentence-level detectors despite perplexity filtering.
Decoupled Search Grounding (DSG) separates search functionality from reasoning models, enabling vendor-agnostic, tunable, and reusable search grounding. DSG achieves near-native accuracy on SimpleQA with 91% lower search cost and 99.4% warm-cache hit rate, while reducing latency by 68% and preserving concise output contracts.
GraphPO introduces a directed acyclic graph framework to represent reasoning rollouts, merging semantically equivalent paths to reduce redundant exploration. It assigns efficiency and correctness advantages to edges, improving inference efficiency and process supervision while reducing advantage-estimation variance. Experiments show GraphPO outperforms chain- and tree-based methods on three LLMs across reasoning and agentic search tasks under identical token or response budgets.
ProfiLLM introduces an agentic LLM pipeline that extracts behavioral signals from ride-hailing logs to generate user profiles. It achieves up to +6.14% relative AUC improvement and up to +4.35% GMV gain in dispatching simulations, with consistent online A/B test results showing +0.47% GMV, +0.33% Completion Rate, and -0.82% Cancel-Before-Accept rate improvements.
SC-GRPO uses per-token KL divergence from self-conditioned trajectories to weight gradients in reinforcement learning. It outperforms GRPO by 8.1% and DAPO by 5.9% across math, code, and agentic tasks, with superior out-of-distribution performance and better results than OPD.
A data-centric approach improves long-context reasoning in large language models, using eight curated datasets with 14K examples across retrieval, multi-evidence synthesis, and reasoning tasks. When paired with minimal outcome-based GRPO training, it achieves average gains of +7.2 to +6.4 points on seven benchmarks, outperforming prior RL training sets, and enhances agentic performance by +4.8 and +7.0 points on GAIA and BrowseComp respectively.
Skill-MAS introduces a new approach that decouples experience retention from parametric updates by modeling orchestration as an evolvable Meta-Skill. It uses a closed-loop process involving multi-trajectory rollouts and selective reflection to distill reusable strategy principles, achieving strong performance gains and robust transferability across tasks and LLMs.
WorldLines introduces a project-driven benchmark for long-horizon embodied household assistance, capturing extended household traces with dialogues, actions, and state changes. It enables evidence-linked samples for Memory QA and Embodied Task Planning, and proposes ObsMem, an observer-grounded memory framework that supports visibility-aware memories and state-aware decisions. Experiments highlight challenges in partial observability and memory translation, with ObsMem providing a stronger reference architecture for such settings.
SGCD introduces an iterative framework to improve GUI agents by addressing supervision gaps in off-trajectory states. It extracts skills from both successful and failed rollouts, using them to guide policy continuations that are mixed with expert trajectories. On OSWorld-Verified, SGCD boosts success rates of three base models from low-30\% to over 50\%.
Decoupled Search Grounding (DSG) separates search functionality from reasoning models, enabling vendor-agnostic, tunable, and reusable search grounding. DSG achieves near-native accuracy on SimpleQA with 91% lower search cost and 99.4% warm-cache hit rate, while reducing latency by 68% and preserving concise output contracts.
RTSGameBench addresses limitations in existing RTS benchmarks by offering diverse gameplay, targeted competency diagnosis, and self-evolving scenario generation. It evaluates vision-language models in strategic reasoning under uncertainty, revealing that state-of-the-art models struggle with multiagent coordination and large-scale tasks.