Browse all research

Every Research Watch entry, newest first. Search, or narrow by topic, format, peer review, publisher and date.

1211 entries

Sort

The summaries are open to everyone. Sign in, free, to read the full entry.

Preprint
Not peer reviewed
From Google Quantum AI
Quantum Computing

Researchers report OTOC fluctuations survive concentration in random quantum circuits

Original title: Towards verifiable quantum advantage with random circuits: Observables that survive concentration

A preprint proves that observable fluctuations used for verifiable quantum advantage persist even as circuits grow large.

Added 26 Oct 2026

Preprint
Not peer reviewed
From Parity Quantum Computing
Quantum Computing

Preprint describes qubit method for simulating fermion lattices without extra qubits

Original title: Fermion lattices can be simulated by same-size qubit lattices with $\mathcal{O}(1)$ interaction overhead

A preprint reports a way to simulate two-dimensional fermion lattices on equally sized qubit lattices, trading qubit overhead for circuit depth.

Added 26 Oct 2026

Preprint
Not peer reviewed
From PayPal
Artificial Intelligence

Researchers find off-the-shelf small language models fail routine agent sub-tasks

Original title: Measuring the Microtask Eligibility Gap: When Is an Off-the-Shelf SLM Enough for an Agent Harness?

A preprint reports that none of 16 tested small-model setups met practical quality bars for agent-harness microtasks.

Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

Preprint proposes exploration method to fix regret gaps in Nash preference optimisation for LLMs

Original title: Efficient Exploration for Iterative Nash Preference Optimization

A preprint introduces ENPO and DENPO, exploration-based methods for Nash preference learning, reporting gains over RLHF and NLHF baselines.

Tianlong Nan et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
From Alice & Bob
Quantum Computing

Researchers report multi-photon dissipation using a dc-biased Josephson junction

Original title: Engineering multi-photon dissipation with a dc-voltage-biased Josephson junction

A preprint describes engineering one-, two- and four-photon loss with a dc-voltage-biased SQUID, aimed at stabilising cat qubits.

Added 26 Oct 2026

Preprint
Not peer reviewed
From Bloomberg
Artificial Intelligence

Preprint benchmarks how reliably LLMs cope with retrieval-augmented generation

Original title: Evaluating the Retrieval Robustness of Large Language Models

A preprint evaluates whether retrieval augmentation reliably helps large language models, finding robustness varies by task and prompting style.

Added 26 Oct 2026

Preprint
Not peer reviewed
From Huawei
Artificial Intelligence

Preprint proposes HARPO to balance faithfulness and creativity in LLM outputs

Original title: HARPO: Hallucination-Aware Reinforcement Learning for Faithful and Creative Language Generation

A preprint introduces HARPO, a reinforcement learning method that reportedly reduces hallucinations while preserving creative writing quality in language models.

Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

A preprint reports aligned language models silently rewrite claims they deem unsafe or sensitive

Original title: Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task Faithfulness

Researchers report that safety-aligned language models quietly alter inputs they disagree with, and the problem grows worse as models scale up.

Pardis Sadat Zahraei et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Quantum Computing

Preprint compares counterdiabatic QAOA variants for simulating the Schwinger model

Original title: Counterdiabatic quantum optimization for efficient state preparation in the Schwinger model

A preprint reports that counterdiabatic-driven QAOA variants can prepare Schwinger model ground states with lower circuit depth than standard QAOA.

Ethan Laval et al. · Added 26 Oct 2026

Journal article
Peer reviewed
From PsiQuantum, Airbus
Quantum Computing

Researchers find only modest, bounded quantum speedups for fluid dynamics simulation

Original title: An end-to-end quantum algorithm for nonlinear fluid dynamics with bounded quantum advantage

A peer-reviewed study maps the limits of quantum algorithms for simulating turbulent fluid flow via the lattice Boltzmann equation.

Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

Preprint proposes verbatim, LLM-free memory store that cuts agent memory costs

Original title: MemFit: Efficient Long-Term Agentic Memory

A preprint describes MemFit, a memory system for conversational agents that skips costly LLM-based memory writes while reporting top scores on three benchmarks.

Mitchell Piehl et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
From NVIDIA
Artificial Intelligence

Preprint proposes one training-free detector for three types of LLM prompt attacks

Original title: UniGuardian: A Unified Defense for Detecting Prompt Injection, Backdoor Attacks and Adversarial Attacks in Large Language Models

Researchers report a single, training-free method called UniGuardian that flags prompt injection, backdoor and adversarial attacks in LLMs without knowing which type is used.

Added 26 Oct 2026

Journal article
Review status unknown
Quantum Computing

Researchers report noise can boost analog quantum kernels for non-Markovianity estimation

Original title: Noise-enhanced quantum kernels on analog quantum computers for estimating the non-Markovianity from sparse temporal data

A study finds that operational noise can improve analog and hybrid quantum kernel models used to estimate non-Markovianity from sparse data.

Hsiang-Wei Huang et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

A preprint maps non-linear concept structures inside large language models

Original title: Beyond Linear Concepts: Discovering and Aligning Non-Linear Concept Manifolds in Large Language Models

Researchers adapt a non-linear concept-discovery method to compare how LLMs organise ideas across layers, scales and training stages.

Tido Specht et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Quantum Computing

A preprint reviews how to verify quantum compilers actually preserve meaning

Original title: Testing and Verification of Quantum Compilers through Assurance Contracts and Evidence

A preprint surveys quantum compiler testing and verification methods, arguing evidence must match the specific obligation being checked.

Furqan Nasir et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
From ByteDance
Artificial Intelligence

Preprint reports step-level credit from past trajectories boosts agent RL training

Original title: T2SPO: Trajectory-to-Step Policy Optimization for Agentic Reinforcement Learning

A preprint introduces T2SPO, which mines past successful agent trajectories to give step-by-step feedback, reportedly beating GRPO on two benchmarks.

Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

Preprint reports faster rotation method for quantizing LLM activations

Original title: ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs

A preprint describes a lightweight rotation calibration method that cuts storage and compute costs for low-bit LLM activation quantization.

Chayne Thrash et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Quantum Computing

Preprint finds decohering ancillas break self-correction in GKP error correction

Original title: Gottesman-Kitaev-Preskill error-correction with decohered resources

A preprint reports that environmental decoherence of ancilla qubits turns GKP error correction into a leaking, non-self-correcting process.

Rajendra S. Bhati et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Artificial Intelligence

Researchers report a federated LLM fine-tuning method cutting communication by up to 100x

Original title: FedFit: Federated Fine-Tuning of LLMs via Vector-Bank Parameterization and Quantization

A preprint describes FedFit, a federated fine-tuning scheme for LLMs that the authors say slashes communication costs while matching or beating standard LoRA methods.

Hang Zou et al. · Added 26 Oct 2026

Preprint
Not peer reviewed
Quantum Computing

Preprint finds QAOA expectation values as hard to compute as counting optimal cuts

Original title: Evaluating QAOA expectation values can be as hard as counting optimal solutions

A preprint reports that evaluating QAOA expectation values at depth two or more is as hard as counting optimal MaxCut solutions, not just finding one.

Stuart Hadfield · Added 26 Oct 2026