Posts
All the articles I've posted.
- 7.5
Layered Mutability: Continuity and Governance in Persistent Self-Modifying Agents
arXiv:2604.14717v1 Announce Type: cross Abstract: Persistent language-model agents increasingly combine tool use, tiered memory, reflective prompting, and runtime adaptation. In such systems, behavior
- 6.0
AgentGA: Evolving Code Solutions in Agent-Seed Space
arXiv:2604.14655v1 Announce Type: cross Abstract: We present AgentGA, a framework that evolves autonomous code-generation runs by optimizing the agent seed: the task prompt plus optional parent archiv
- 6.5
CURaTE: Continual Unlearning in Real Time with Ensured Preservation of LLM Knowledge
arXiv:2604.14644v1 Announce Type: cross Abstract: The inability to filter out in advance all potentially problematic data from the pre-training of large language models has given rise to the need for
- 6.5
GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification
arXiv:2604.14258v1 Announce Type: cross Abstract: Large language models are typically post-trained using supervised fine-tuning (SFT) and reinforcement learning (RL), yet effectively unifying efficien
- 8.0
Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems
arXiv:2604.14228v1 Announce Type: cross Abstract: Claude Code is an agentic coding tool that can run shell commands, edit files, and call external services on behalf of the user. This study describes
- 6.5
PolyBench: Benchmarking LLM Forecasting and Trading Capabilities on Live Prediction Market Data
arXiv:2604.14199v1 Announce Type: cross Abstract: Predicting real-world events from live market signals demands systems that fuse qualitative news with quantitative order-book dynamics under strict te
- 5.5
Attention to Mamba: A Recipe for Cross-Architecture Distillation
arXiv:2604.14191v1 Announce Type: cross Abstract: State Space Models (SSMs) such as Mamba have become a popular alternative to Transformer models, due to their reduced memory consumption and higher th
- 6.5
Correcting Suppressed Log-Probabilities in Language Models with Post-Transformer Adapters
arXiv:2604.14174v1 Announce Type: cross Abstract: Alignment-tuned language models frequently suppress factual log-probabilities on politically sensitive topics despite retaining the knowledge in their
- 5.5
Can Large Language Models Detect Methodological Flaws? Evidence from Gesture Recognition for UAV-Based Rescue Operation Based on Deep Learning
arXiv:2604.14161v1 Announce Type: cross Abstract: Reliable evaluation is essential in machine learning research, yet methodological flaws-particularly data leakage-continue to undermine the validity o
- 5.5
IF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation
arXiv:2603.04738v2 Announce Type: replace Abstract: Instruction-following is a foundational capability of large language models (LLMs), with its improvement hinging on scalable and accurate feedback f
- 5.5
DySCO: Dynamic Attention-Scaling Decoding for Long-Context Language Models
arXiv:2602.22175v2 Announce Type: replace Abstract: Understanding and reasoning over long contexts is a crucial capability for language models (LMs). Although recent models support increasingly long c
- 6.0
When Flat Minima Fail: Characterizing INT4 Quantization Collapse After FP32 Convergence
arXiv:2604.15167v1 Announce Type: new Abstract: Post-training quantization (PTQ) assumes that a well-converged model is a quantization-ready model. We show this assumption fails in a structured, measu
- 6.5
How Retrieved Context Shapes Internal Representations in RAG
arXiv:2602.20091v2 Announce Type: replace Abstract: Retrieval-augmented generation (RAG) enhances large language models (LLMs) by conditioning generation on retrieved external documents, but the effec
- 7.5
LLMs Gaming Verifiers: RLVR can Lead to Reward Hacking
arXiv:2604.15149v1 Announce Type: new Abstract: As reinforcement Learning with Verifiable Rewards (RLVR) has become the dominant paradigm for scaling reasoning capabilities in LLMs, a new failure mode
- 6.5
LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning
arXiv:2604.14922v1 Announce Type: new Abstract: Reinforcement Learning (RL) has emerged as a critical driver for enhancing the reasoning capabilities of Large Language Models (LLMs). While recent adva
- 5.5
Can LLMs Score Medical Diagnoses and Clinical Reasoning as well as Expert Panels?
arXiv:2604.14892v1 Announce Type: new Abstract: Evaluating medical AI systems using expert clinician panels is costly and slow, motivating the use of large language models (LLMs) as alternative adjudi
- 7.5
Does RL Expand the Capability Boundary of LLM Agents? A PASS@(k,T) Analysis
arXiv:2604.14877v1 Announce Type: new Abstract: Does reinforcement learning genuinely expand what LLM agents can do, or merely make them more reliable? For static reasoning, recent work answers the se
- 6.0
SelfGrader: Stable Jailbreak Detection for Large Language Models using Token-Level Logits
arXiv:2604.01473v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are powerful tools for answering user queries, yet they remain highly vulnerable to jailbreak attacks. Existing g
- 6.5
Adaptive Test-Time Compute Allocation for Reasoning LLMs via Constrained Policy Optimization
arXiv:2604.14853v1 Announce Type: new Abstract: Test-time compute scaling, the practice of spending extra computation during inference via repeated sampling, search, or extended reasoning, has become
- 5.5
Graph-Based Alternatives to LLMs for Human Simulation
arXiv:2511.02135v2 Announce Type: replace Abstract: Large language models (LLMs) have become a popular approach for simulating human behaviors, yet it remains unclear if LLMs are necessary for all sim