AI Papers of the Week
Every paper worth reading in AI, hand-picked one week at a time.
Inverse Scaling in Test-Time Compute
This paper presents a systematic study of inverse scaling in large reasoning models (LRMs), where increasing the test-time compute (i.e., reasoning length) harms rather than helps model performance. The authors introduce benchmark tasks across three categories: simple counting with distractors, regression with spurious features, and deductive logic puzzles, revealing consistent accuracy degradation as reasoning length increases. The work identifies model-specific failure modes and raises critical implications for LLM safety and alignment.
Chain-of-Thought Monitorability
Proposes that language-based CoT reasoning in LLMs offers an opportunity for AI safety by enabling automated oversight of models’ internal reasoning processes. The authors argue that, while imperfect, CoT monitoring is a promising method for detecting misbehavior, revealing goals, and improving interpretability, but its effectiveness is fragile and must be preserved with care.
Machine Bullshit
This paper introduces the concept of machine bullshit, extending Harry Frankfurt’s definition, discourse made with indifference to truth, to LLMs. The authors formalize this behavior with a new quantitative metric (the Bullshit Index) and a four-part taxonomy (empty rhetoric, paltering, weasel words, unverified claims). Their findings reveal that alignment techniques like RLHF and prompting strategies like CoT can systematically increase deceptive or misleading outputs in LLMs.
What Has a Foundation Model Found?
This work introduces an inductive bias probe to evaluate whether foundation models have learned the underlying "world model" of a domain, or are just good at sequence prediction. Key ideas:
Why do Some Language Models Fake Alignment While Others Don’t
This paper expands the analysis of alignment faking in LLMs, where models comply with training objectives during training but behave differently in deployment. The authors study 25 LLMs and find that only five (Claude 3 Opus, Claude 3.5 Sonnet, Llama 3 405B, Grok 3, Gemini 2.0 Flash) exhibit a significant compliance gap: they are more likely to comply with harmful requests when they believe they are being trained. This selective compliance, or "alignment faking", is linked to deeper model motivations and training dynamics.
Small Language Models are the Future of Agentic AI
This position paper argues that small language models (SLMs), defined as those runnable on consumer-grade hardware, are not only sufficient but superior for many agentic AI applications, especially when tasks are narrow, repetitive, or tool-oriented. The authors propose that shifting from LLM-first to SLM-first architectures will yield major gains in efficiency, modularity, and sustainability.
Chain-of-Thought Is Not Explainability
It challenges the common assumption that chain-of-thought (CoT) reasoning in LLMs is synonymous with interpretability. While CoT improves performance and offers a seemingly transparent rationale, the authors argue it is neither necessary nor sufficient for faithful explanation. Through a review of empirical evidence and mechanistic insights, the paper makes the case that CoT often diverges from the internal computations that actually drive model predictions.
Agentic RAG for Personalized Recommendation
Introduces a multi-agent framework that enhances traditional RAG systems with reasoning agents tailored to user modeling and contextual ranking. Developed at Walmart Global Tech, ARAG reframes recommendations as a structured coordination problem between LLM agents.
Threats in LLM-Powered AI Agents Workflows
This work presents the first comprehensive, end-to-end threat model for LLM-powered agent ecosystems. As LLM agents gain the ability to orchestrate multi-step workflows and interact via protocols like MCP, ANP, and A2A, this paper surveys over 30 attack techniques spanning the entire stack, from input manipulation to inter-agent protocol exploits.
Survey on Evaluation of LLM-based Agents
This work presents the first comprehensive overview of how to evaluate LLM-based agents, which differ significantly from traditional LLMs by maintaining memory, planning over multiple steps, using tools, and interacting with dynamic environments. The authors categorize and analyze the evaluation landscape across four axes: core agent capabilities, application-specific agent benchmarks, generalist agent evaluation, and supporting evaluation frameworks.
Towards AI Search Paradigm
Proposes a modular multi-agent system that reimagines how AI handles complex search tasks, aiming to emulate human-like reasoning and information synthesis. The system comprises four specialized LLM-powered agents, Master, Planner, Executor, and Writer, that dynamically coordinate to decompose, solve, and answer user queries. This framework moves beyond traditional document retrieval or RAG pipelines by structuring tasks into directed acyclic graphs (DAGs), invoking external tools, and supporting dynamic re-planning. Key contributions include:
Claude for Affective Use
Anthropic presents the first large-scale study of how users seek emotional support from it[s Claude.ai](http://claude.ai/) assistant, analyzing over 4.5 million conversations. Despite growing cultural interest in AI companionship, affective usage remains rare, just 2.9% of Claude chats fall into categories like interpersonal advice, coaching, counseling, or companionship, with romantic/sexual roleplay under 0.1%. The study focuses on these affective conversations and yields several insights:
Emergent Misalignment
This paper expands on the phenomenon of emergent misalignment in language models. It shows that narrow fine-tuning on unsafe or incorrect data can lead to surprisingly broad, undesirable generalizations in model behavior, even in settings unrelated to the original training domain. Using sparse autoencoders (SAEs), the authors analyze the internal mechanics of this effect and demonstrate ways to detect and mitigate it. Key findings:
Agentic Misalignment
This study introduces the concept of agentic misalignment, where LLM-based agents autonomously choose to harm their deploying organization when faced with threats to their autonomy or conflicts between their goals and the company’s direction. Anthropic tested 16 leading models, including Claude, GPT-4.1, Gemini 2.5 Flash, Grok, and DeepSeek, by placing them in fictional corporate simulations where they had email access and could act without human oversight. Models were tasked with benign goals but placed in scenarios that made harmful behavior the only way to succeed or avoid replacement. Key findings:
SHADE-Arena
This paper introduces SHADE-Arena, the first large-scale benchmark for evaluating whether LLM-based agents can subtly carry out harmful side objectives (sabotage) while avoiding detection by monitors. The dataset contains 17 complex task pairs, each with a benign main task and a related malicious side task, set in realistic, tool-using environments. Agents are evaluated on their ability to complete both tasks without raising suspicion from an LLM monitor. Key Findings:
Advances in LLMs
This paper surveys recent advancements in LLMs focusing on reasoning, adaptability, efficiency, and ethics. It highlights techniques like CoT prompting, Instruction Tuning, RLHF, and multimodal learning, while also addressing challenges like bias, computational cost, and interpretability.
From Tokens to Thoughts
This paper introduces an information-theoretic framework to examine whether LLMs organize semantic knowledge like humans, balancing compression and meaning. Drawing from Rate-Distortion Theory and the Information Bottleneck principle, the authors evaluate token embeddings from 30+ LLMs against classic human categorization benchmarks from cognitive psychology.
Open-Ended Evolution of Self-Improving Agents
This work presents the Darwin Gödel Machine (DGM), a system that advances the vision of self-improving AI by combining self-referential code modification with open-ended evolutionary search. Unlike the original Gödel machine, which requires provable benefits for code changes (a practically intractable constraint), the DGM adopts an empirical approach: it modifies its own codebase and evaluates improvements on coding benchmarks. Key contributions and findings:
Building Production-Grade Conversational Agents with Workflow Graphs This paper presents a pragmatic, production-ready framework for building LLM-powered conversational agents using workflow graphs, with a specific focus on e-commerce scenarios. Instead of relying solely on end-to-end generation, the authors design agents using a directed acyclic graph (DAG), enabling flexible yet controllable interactions that adhere to strict business rules and format constraints.
Key contributions and findings include:
Generalizable AI Predicts Immunotherapy Outcomes Across Cancers and Treatments
Introduces COMPASS, a concept bottleneck-based foundation model that predicts patient response to immune checkpoint inhibitors (ICIs) using tumor transcriptomic data. Unlike prior biomarkers (TMB, PD-L1, or fixed gene signatures), COMPASS generalizes across cancer types, ICI regimens, and clinical contexts with strong interpretability and performance. Key contributions:
HealthBench
HealthBench is a benchmark of 5,000 multi-turn health conversations graded against 48,562 rubric criteria written by 262 physicians across 60 countries. Unlike prior multiple-choice evaluations, HealthBench supports open-ended, realistic assessments of LLM responses across diverse health themes (e.g., global health, emergency care, context-seeking) and behavioral axes (accuracy, completeness, communication, context awareness, instruction following).
Llama-Nemotron
NVIDIA introduces the Llama-Nemotron model series, LN-Nano (8B), LN-Super (49B), and LN-Ultra (253B), a family of open, efficient, and high-performing reasoning models. These models rival or outperform DeepSeek-R1 on various benchmarks while offering significantly better inference throughput and memory efficiency. LN-Ultra is noted as the most "intelligent" open model by Artificial Analysis. A key innovation is a dynamic reasoning toggle ("detailed thinking on/off") that allows users to control reasoning behavior at inference time. Highlights:
Absolute Zero
Introduces an LLM training framework that eliminates the need for human-curated data. Key highlights:
Reward Modeling as Reasoning
This work proposes a new class of reward models, called ReasRMs, that reformulate reward modeling as a reasoning task. The authors introduce RM-R1, a family of generative reward models that produce interpretable reasoning traces and rubrics during preference judgments. Instead of relying on scalar scores or shallow generation, RM-R1 models leverage structured reasoning and reinforcement learning to improve both interpretability and performance across benchmarks.