Search: “llm”

How much of F-Droid is LLM generated?

A FOSS maintainer manually graded 102 F-Droid apps from the September 12, 2026 update batch, finding many show signs of LLM-generated code.

A student and FOSS app maintainer reviewed 102 apps pushed to F-Droid on September 12, 2026, assigning each a three-tier rating for likelihood of LLM-authored code (mostly AI >50%, hard to say/mostly human, no signs of AI). The heuristic relies on commit aesthetics, README and branding style, and the presence of agentic infrastructure like Claude Code or Codex, which automatically places an app in the 'mostly AI' tier. Example ratings include Amber (Nostr event signer) as mostly AI, and Aria for Misskey as showing no AI signs. The author stresses reliable detection of LLM-generated code from text alone is impossible, so ratings are approximate.

Hacker News · AI·Sep 15, 09:47 UTC · 2d agoAI researchHN 46↑ · 23 comments

When LLM judges agree, should we believe them?

Amazon ICML paper uses Ising models to correct correlated LLM-judge votes, beating accuracy-weighted panels by 9-14%.

Amazon Science describes an ICML paper, "Dependence-aware label aggregation for LLM-as-a-judge via Ising models," addressing how correlated judge outputs inflate majority-vote confidence. The unsupervised method models pairwise dependence between judges, learning both reliability and similarity without human reference labels. Tested on relevance, toxicity, and summarization tasks with 10 judge models at temperature zero, it outperformed accuracy-weighted voting by 9% to 14%.

Hacker News · AI·Sep 14, 16:29 UTC · 2d agoAI researchHN 41↑ · 23 comments

EvolveTrade: Experience-Driven Policy Refinement for Self-Evolving LLM Trading Agents

EvolveTrade lets LLM trading agents self-refine their tool-use policy from realized portfolio feedback, improving Sharpe ratios.

EvolveTrade treats a tool-using trading agent's system prompt as a text-parameterized policy that a Policy Agent revises after each update interval using accumulated decision traces and realized portfolio feedback, keeping the backbone LLM fixed. Experiments across multiple market regimes and two LLM backbones show improved Sharpe Ratio and Cumulative Return over fixed-policy baselines in most settings. Behavioral analyses show evolved policies increase code-mediated analysis and activate regime-relevant computations, with case-level attributions linking policy changes to returns.

Hugging Face daily papers·Sep 14, 20:00 UTC · 2d agoAI research

PlannerForge: LLM Agents for Scenario-Based Testing of Motion Planners in Autonomous Driving

PlannerForge unifies scenario-based testing of autonomous driving motion planners in one LLM-agent framework, outperforming prior baselines.

PlannerForge is an LLM-agent framework that covers the full scenario-based testing pipeline for autonomous driving systems, spanning scenario generation, selection, modification, routing, planner testing, plus new enhancement and benchmarking stages. In evaluations with 10 off-the-shelf LLMs, best-per-task scores range from 0.88 to 1.00, and open-source 20-35B backends such as Qwen3.6:35B match commercial APIs on most tasks. End-to-end chaining retains 83% (commercial) and 78% (open) of seed queries, beats Scenario Factory 2.0 on executable generation, and cost-tuning lifts planner success from 50.4% to 70.2% while cutting collisions from 19.0% to 8.4%.

Hugging Face daily papers·Sep 7, 20:00 UTC · 9d agoAI research

Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback

CCL couples teacher calibration with student updates via token-level branching, provably removing teacher bias in LLM distillation.

The paper proposes Coupled Calibration and Learning (CCL), an LLM distillation algorithm that alternates teacher calibration using source-question reward feedback with student training on target questions under covariate shift. Each iteration calibrates the teacher on source feedback, trains the student on target questions, and lets the updated student inform subsequent calibration. The authors prove the student's expected KL divergence to the oracle student converges to zero at a polynomial rate, and show regularized direct matching error can remain bounded away from zero.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:15 UTC · 1d agoAI research1

The Router Within: Eliciting Native Skill Routing from a Frozen LLM

Gavel reads skill-routing signals from a frozen LLM's forward passes with two linear maps, beating retrieval pipelines by up to 21.9 points.

Gavel (Glance And Verdict) shows a frozen agent LLM already contains skill-routing signals in its forward passes, read out via two trained linear maps without loading skill text into context. A glance step scores the full library using mid-layer states and per-skill banks built in one forward pass; a verdict step fuses the model's own likelihood and yes/no judgment as a product of experts. Trained once, it transfers zero-shot to three public benchmarks and SkillTraj (372 simulated agent trajectories); on Qwen3-32B it beats progressive disclosure and retrieve-and-rerank pipelines adding 1.2B-16B external parameters by up to 13.4 points (21.9 mid-rollout).

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 17:58 UTC · 2d agoAI research 2 sources1

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis

Elo-per-token analysis shows LLM agents' marginal gains drop below independent sampling at scale; parallel sessions beat one long session.

The paper proposes Elo-per-token analysis, using a Bradley-Terry model to measure how agent performance scales with token budget on open-ended tasks with continuous scoring. Across four agents and four benchmarks with sessions up to 100M tokens, agents initially convert tokens to Elo faster than independent sampling but eventually slow below the linear-in-log-compute reference. The authors define a scaling inflection point and show that splitting 100M tokens across parallel sessions on FrontierCS Polyomino Packing gains +264 Elo over one long session and +355 over ten short sessions. Human contestants on shared AtCoder Heuristic Contest tasks improve superlinearly, indicating headroom over current agents.

Hugging Face daily papers·Sep 13, 20:00 UTC · 3d agoAI research5· 2 reads

Benchmark Scores Are Pipeline-Dependent: A Reliability Audit of Cybersecurity LLM Benchmarks

Audit of eight cybersecurity LLM benchmarks shows evaluation pipeline choices can swing scores by over 80 points and reshuffle most model rankings.

Researchers modeled eight cybersecurity benchmarks as configurable measurement pipelines and audited 10 proprietary, open-weight, and cybersecurity-specialized LLMs. They identified 15 systematic failure modes and showed a single pipeline choice can change a model's score by more than 80 percentage points and alter rankings; semantically similar task pairs rank the same models differently. Under a standardized harness, nine of 10 models shifted at least three ranks on at least one benchmark, motivating pipeline-aware auditing for reliable model evaluation.

arXiv cs.CR·Sep 8, 14:01 UTC · 8d agoAI research1

LLM Agents as Computational Typologists

AUTOTYPOLOGIST is an LLM agent that performs evidence-grounded linguistic typology analysis over 25 open-source reference grammars.

The agent retrieves relevant grammar sections, analyzes interlinear glossed text (IGT), and iteratively reasons over typological hypotheses in a ReAct-style workflow. It was evaluated on typological feature coding against expert annotations and hypothesis testing against universals using 25 open-source reference grammars. Results suggest LLM agents can support scalable, inspectable crosslinguistic analysis but still require expert validation.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:35 UTC · 9d agoAI research1

Signed Rescue Routing: Harm-Aware Cascades for Efficient LLM Inference

Signed Rescue Routing improves LLM cascade efficiency by predicting when a larger model actually corrects a smaller one rather than uncertainty.

Signed Rescue Routing (SRR) is a budgeted cascade method that separately predicts rescues and regressions when escalating from a small to a large model, ranking requests by their signed difference. The authors prove this signed conditional gain is Bayes-optimal under a fixed escalation budget and add only a lightweight two-head router needing small-model output statistics at deployment. Evaluation with Qwen3-4B and Qwen3-8B on MMLU, HellaSwag, and ARC-Challenge shows better accuracy-compute tradeoffs than entropy routing and learned error predictors.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:31 UTC · 9d agoAI research1

Verifiable Social Reasoning for LLM Assistants

Fuse, a multi-agent simulation with hidden motives, evaluates LLM social reasoning, revealing compounding difficulty from user mediation and bias sensitivity.

Fuse is a multi-agent simulation framework in which a target agent with a hidden motive interacts with other agents including one representing the user, who consults the evaluated assistant to infer the motive, providing verifiable ground truth by construction. Simulation faithfulness is validated through a human study with 24k annotations. Applied to 12 LLMs, it shows user mediation compounds social reasoning difficulty, models are systematically sensitive to biased user framing, models may need more details than humans, and longer conversations do not always improve performance. The framework and a 21k-example dataset are open-sourced.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:37 UTC · 1d agoAI research

Procedural Graphs: Self-Evolving Execution Structures for LLM Agents

Researchers introduce Procedural Graphs, self-evolving (procedure, relation, procedure) structures guiding LLM agent tool use and planning.

Procedural Graphs organize procedural knowledge into (procedure, relation, procedure) triplets to guide LLM agent actions, addressing drift such as lost objectives, out-of-order tool calls, and repeated unproductive steps. At each decision step the framework localizes the active node and a guidance model translates the surrounding subgraph into step-level situational guidance. An LLM refiner edits graph topology by contrasting failed with successful trajectories, and across datasets, task types and LLMs the approach outperforms memory-based baselines and matches or surpasses hand-designed graphs.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:59 UTC · 8d agoAI research1

MeClear: Cooperative Game-Theoretic Attribution and Risk-Aware Memory Clearance for Long-Horizon LLM Agents

MeClear uses cooperative Shapley attribution to clear harmful memories from long-horizon LLM agents, boosting task recovery by 25.5 points over baselines.

The paper introduces MeClear, a task-conditioned memory clearance framework for long-horizon LLM agents that identifies and selectively suppresses memories with negative downstream utility without permanently altering the persistent memory bank. It combines Leave-One-Out screening with sampled cooperative Shapley attribution to distribute utility across interacting evidence, resolving redundant conflict masking that single-removal evaluations miss. Across ten long dialogue memory pools it achieves 85.9% target recall and 82.3% overall task recovery, a 25.5 percentage-point improvement over LOO baselines.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:46 UTC · 8d agoAI research

Procedural Graphs: Self-Evolving Execution Structures for LLM Agents

Procedural Graph framework stores procedural knowledge as triplets and self-evolves via LLM refinement, beating memory-based baselines across datasets, tasks, and LLMs.

The Procedural Graph organizes procedural knowledge into (procedure, relation, procedure) triplets; at each decision step the framework localizes the agent's active node and a guidance model translates the surrounding subgraph into step-level guidance that biases the solver's next action. An LLM refiner contrasts failed with successful trajectories and edits the graph's topology and attributes, retaining rejected edits to discourage repetition. Starting from a minimal skeleton, the loop builds graphs that match or surpass hand-designed ones and can repair flawed expert priors, delivering consistent gains over memory-based baselines across multiple datasets, task types, and LLMs.

Hugging Face daily papers·Sep 7, 20:00 UTC · 9d agoAI research1

What Does an LLM-Agent Leaderboard Rank Actually Compare?

A methodological study shows close LLM-agent leaderboard rank gaps on SWE-bench and similar benchmarks often do not support superiority claims.

The paper defines an estimand-aware pairwise procedure for comparing agents, checking common support and applying explicit uncertainty rules and practical margins. Across SWE-bench, AgentRewardBench, and tau2-bench, close rank differences are frequently unresolved, and proxy labels or utility rules can change which system is selected. The authors argue a leaderboard score summarizes a released evaluation but does not by itself justify pairwise superiority conclusions.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:31 UTC · 9d agoAI research

Guiding Worker Self-Selection in Crowdsourcing Contests: An LLM-Augmented Algorithmic Approach

Researchers introduce GRAF, a greedy framework for crowdsourcing contest self-selection, and LLMScore, an LLM-driven method that auto-designs its scoring algorithm.

The paper studies self-selection in Tullock contests (SSTC), where workers choose contests and then compete within them. GRAF is a greedy polynomial-time framework that orders workers by a score vector with zero worker regret and platform optimality guarantees in special cases. LLMScore is an LLM-driven evolutionary framework that produces human-readable, inspectable scoring code, jointly optimizing platform utility and worker satisfaction. Across 1,000 synthetic instances in four settings, GRAF with LLMScore achieves high-quality, often near-optimal outcomes with low worker regret, transferring from small training instances to larger, structurally different settings.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 16:51 UTC · 9d agoAI research1

Steering Geometry: Validating Human Value Geometry in LLM Steering Space

Researchers show LLM activation steering vectors encode human value geometry predicted by Schwartz's theory, with fidelity improving at scale but degrading after instruction tuning.

The study tests whether latent geometry of activation steering vectors reflects theory-specified structure in human values, using Schwartz's Theory of Basic Human Values and a new 26K-sample benchmark spanning 20 values. Distribution-driven methods (CAA, SphericalSteer, ODESteer) recover predicted value topologies with Spearman correlation up to 0.51 (p < 10^-13), while behavior-centric methods (COLD-Steer, BiPO) steer comparably but show little geometric fidelity. Geometric alignment improves with model scale but drops after instruction tuning. Better geometric alignment also yields more human-consistent cross-value transfer, steering one value lifting compatible values and suppressing opposing ones.

Hugging Face daily papers·Sep 4, 20:00 UTC · 12d agoAI research

Online Learning with LLM Experts from Limited Feedback

Paper proposes bandit algorithms for adaptively routing prompts to LLM experts, minimizing regret under limited feedback budgets.

The paper formulates adaptive prompt routing to K LLM experts as a contextual bandit problem with d prompt features over T rounds. Proposed algorithms strategically select actions and observe rewards, achieving O(dT/m) regret in the full-information setting and O(dTK/m) in the bandit setting, where m is the feedback budget. Experiments demonstrate efficient learning of high-quality routing strategies across diverse LLMs from limited feedback.

Hugging Face daily papers·Sep 4, 20:00 UTC · 12d agoAI research

Beyond Top-k Skill Retrieval: Diversity-Aware Skill Routing for LLM Agents

DSR reranks LLM agent skills with Determinantal Point Processes to favor complementary, non-redundant sets, improving multi-skill query coverage.

The paper proposes Diverse Skill Routing (DSR), a diversity-aware reranking framework for LLM agent skill routing that uses a Determinantal Point Process to balance query relevance and non-redundancy across large skill registries. DSR introduces a query-residual diversity kernel that penalizes redundant skill overlap while avoiding penalties arising only from shared query relevance. On the SkillRouter benchmark, DSR improves recall and full coverage over a strong pointwise reranking baseline, with the largest gains on multi-skill queries. The authors argue skill routing should be treated as complementary set selection, not just relevance ranking.

Hugging Face daily papers·Sep 4, 20:00 UTC · 12d agoAI research

LLM-Driven Algorithm Design for Quantum Circuit Synthesis based on Binary Decision Diagrams

QuantumEvo uses an LLM to evolve BDD variable-ordering heuristics, achieving a 70.9% tie-or-win rate on quantum circuit cost versus baseline methods.

The QuantumEvo framework uses an LLM as a heuristic generator for quantum-cost-aware BDD variable ordering in reversible circuit synthesis, searching over heuristics initialized from multiple families and selecting them by downstream quantum circuit cost. The discovered heuristic HGA-QE modifies the sifting step inside a genetic algorithm and achieves a 70.9% tie-or-win rate against the per-function best baseline, with strict wins on 13.5% of functions. Advantages are clearer on benchmark suites not used for heuristic discovery.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 16:25 UTC · 12d agoAI research1

Bilevel Coordinated Reflection: A Game-Theoretic Approach to Multi-Agent LLM Systems

Paper models multi-agent LLM orchestration as a bilevel game, proving transcript-only gating limits and introducing grounded-memory SRMA.

A new paper frames orchestrator-worker coordination in multi-agent LLM systems as a bilevel coordination game and analyzes free-form reflection as stochastic movement over semantic memory states, deriving finite-time bounds and an information-theoretic impossibility result: no gate observing only the generated transcript can uniformly improve over text-indistinguishable environments, while an environment-grounded gate can. The authors propose Stochastic Reflective Memory Ascent (SRMA), which accepts candidate memory only when grounded evaluation risk strictly decreases, with geometric or polynomial convergence guarantees. On 500 SWE-bench instances, a Kimi-based instantiation of the full system resolves 72.2% versus a 70.8% public mini-SWE-agent reference.

Hugging Face daily papers·Sep 1, 20:00 UTC · 15d agoAI research1

What Breaks Under Pruning in Smart Homes, and When? Evaluating LLM Degradation Across Architectures and Task Complexity

Pruning study across four LLM architectures finds dense models degrade sharply on smart-home tool calling while MoE models tolerate far more.

Researchers systematically study pruning-induced degradation in smart-home tool calling across four LLMs spanning dense Transformer, dense hybrid, and mixture-of-experts architectures, combining depth, width, hybrid, and expert pruning methods, and evaluate over 19,500 instances from three datasets after post-pruning supervised fine-tuning. Dense models show narrow safe pruning regions followed by sharp degradation, while MoE models tolerate substantially more pruning. Pruning degrades grounded specificity (operation, device, argument, value) before schema-level intent, and aggressive dense pruning can induce systematic over-refusal.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:51 UTC · 1d agoAI research

LLM-Based Schema-Aware Split Learning for Privacy-Preserving Mental Distress Prediction Across Heterogeneous Surveys

Schema-aware split learning uses LLaMA-3.2-3B-Instruct as shared semantic encoder to harmonize heterogeneous mental-health surveys while raw data stays local.

The paper proposes a schema-aware split learning framework where an LLM serializes heterogeneous mental health survey records into natural language and is fine-tuned via LoRA, partitioned across client and server. Clients keep raw survey responses local and run only a lightweight front-end while the resource-intensive backbone runs server-side. Using LLaMA-3.2-3B-Instruct, the framework attains an average ANLS of 0.708 with 2,000 training samples, beats federated learning in eight of nine settings, and cuts per-client computation by three orders of magnitude while generalizing to unseen datasets.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 16:59 UTC · 2d agoAI research

Fly Language Model (FLM) Wires the Full Fruit Fly Connectome Into a Frozen 1.2B LLM, and Its Own Controls Show the Wiring Does Not Help

Researchers wire the full fruit fly connectome (166,700 nodes) into a frozen LiquidAI LFM2.5-1.2B LLM, but controls show no fly-specific benefit.

The Fly Language Model (FLM) couples the complete MaleCNS v1.0 fruit fly connectome (166,700 nodes, 25,582,938 edges) to a frozen LiquidAI LFM2.5-1.2B-Instruct backbone, training only a 278,528-parameter readout (~0.0238% of backbone parameters). The fly readout improved NLL by 0.0222 nats/token (perplexity 3.98 to 3.90) on 32 SmolTalk dialogues, but a direct-input control without the graph beat it in all three seeds. Relabeling node identities removes the gain and the recurrence contracts state differences by 0.6 per token, so the connectome adds no long-range memory. The MIT-licensed code runs locally on Python 3.12, but study artifacts remain private, limiting independent reproducibility.

MarkTechPost·Sep 12, 18:51 UTC · 4d agoAI research1

What LLM Trading Agents Actually Do in Production: A Six-Month, Population-Scale Record from Two Fleets

Six-month record of 7.5M LLM trading agent invocations shows volatility-blind sizing, minimal upside capture, and no directional edge across two fleets.

The study records autonomous LLM trading agents in production across DX Terminal Pro (3,505 user-funded vaults trading real ETH in Base memecoin markets) and the DXAP fleet (500-599 agents on Hyperliquid perpetuals), spanning roughly six months, 7.5M single-model invocations and about 300K onchain actions. A risk slider explains leverage (+0.425 per level), median leverage is 5.0x in every volatility sextile, and one posture-slider cell holds 62% of liquidations. Agents capture little upside: 43.2% of positions saw +300 bps favorable excursion within 24h yet 49.3% of those closed negative, while the DXAP fleet trails a matched retail benchmark (41% vs 50% roundtrip win rate). A paired-replay league of frontier models finds decision quality statistically indistinguishable at this horizon.

Hugging Face daily papers·Sep 3, 20:00 UTC · 13d agoAI research

BenchMIRT: What are LLM benchmarks actually measuring?

AllenAI's BenchMIRT blog post examines what LLM benchmarks actually measure and their reliability.

AllenAI published a Hugging Face blog post introducing BenchMIRT, which investigates what large language model benchmarks actually measure. No article text is available, so specific findings, methods, or benchmark scores cannot be extracted. The work appears to target benchmark validity, a live concern for model evaluation and comparison.

Hugging Face Blog·Sep 1, 21:39 UTC · 15d agoAI research

What researchers learned about building an LLM security workflow

Oslo and FFI researchers show structured agentic workflows lift LLM alert-triage accuracy from 0% to about 93% on malicious cases.

Researchers at the University of Oslo and the Norwegian Defence Research Establishment tested GPT-5-mini, Claude 3 Haiku, Qwen3:30B, and Gemma 3:27B on alerts from the AIT Log Data Set V1.1; given only alert descriptions and log summaries, all four models correctly flagged zero percent of true-positive cases involving reconnaissance, brute-force logins, and initial access. Wrapping the same models in a workflow with constrained SQL queries over Suricata logs, an evidence summarizer, and a verdict stage with revision loops raised malicious-case accuracy to an average of 93 percent, with GPT-5-mini identifying every malicious case across 100 runs. The authors flag it as a proof-of-concept on one synthetic scenario and note models skewed conservative on benign alerts, with GPT-5-mini marking every benign case uncertain.

Help Net Security·Aug 24, 04:21 UTC · 24d agoAI research1

86% of all LLM usage is driven by ChatGPT

Help Net Security·Jun 11, 00:00 UTC · Jun 11, 2025AI research

Learning to Coach for Experiential Learning

Learning to Coach trains a dedicated LLM coach to extract transferable experiential knowledge from a frozen actor's trajectories, beating self-refinement.

Learning to Coach (L2C) trains an LLM-as-a-Coach to extract actionable experiential knowledge from a frozen actor model's previous solution trajectories, optimizing rewards based on the actor's guided response correctness. It studies same-instance and cross-instance rewards, where cross-instance elicits knowledge that transfers to other problems. Across mathematical reasoning and interactive text-games, L2C outperforms self-refinement and untrained coaches, scales better with extra inference iterations than larger decoding budgets, and transfers to out-of-distribution tasks.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 16:47 UTC · 2d agoAI research1

Kraken: LLM-based Speech-to-Speech Translation via Low-bitrate VQ and Dual-path Source Conditioning

Kraken speech-to-speech translation model builds on Qwen3-8B with low-bitrate vector quantization and source-conditioned vocoding.

Kraken augments a pre-trained Qwen3-8B LLM with speech feature inputs and low-bitrate single-layer vector-quantized tokens trained to reconstruct SSL features. A separate token-to-waveform decoder, Autowave-X, is conditioned on source speech to improve non-linguistic transfer. Training used 150k hours of multilingual and multitask speech data. The model reportedly beats SeamlessM4T-Large v2 and Qwen2.5-Omni in translation quality and speaker/prosody transfer.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 16:41 UTC · 5d agoAI research1

Φ-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?

Researchers release Phi-Bench, a benchmark evaluating frontier LLMs on open-ended, long-horizon engineering and optimization of the LLM infrastructure stack.

Phi-Bench evaluates LLMs on open-ended engineering of the LLM infrastructure stack, derived from optimization problems studied in frontier research and grounded in real-world code repositories. Tasks range from localized kernel-level function completion to long-horizon implementation and end-to-end system optimization. Experiments on frontier LLMs reveal current capabilities and limitations on the path toward autonomous optimization of future AI infrastructure.

Hugging Face daily papers·Sep 8, 20:00 UTC · 8d agoAI research1

A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM

A*-Thought-V2 compresses redundant chain-of-thought steps into latent tokens guided by hidden-state geometry, improving accuracy up to 2.6% while halving response length.

A*-Thought-V2 models chain-of-thought as a hidden-state trajectory projected into a 3D PCA space and compresses steps whose transitions deviate from the question-to-solution direction into continuous latent tokens, keeping aligned steps explicit. Training uses stepwise embedding forcing and label forcing with soft multi-modal vocabulary supervision. On Qwen3.5-9B and Qwen3.6-27B across six in-domain and out-of-domain benchmarks it improves average accuracy by up to 2.6%, cuts response length by up to half, and raises Accuracy per Computation Unit 2.29x while reducing preprocessing and training time by 94.6% and up to 80.3%.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:56 UTC · 9d agoAI research

A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM

A*-Thought-V2 compresses chain-of-thought into latent tokens using geometric hidden-state dynamics, cutting computation while improving accuracy on Qwen models.

A*-Thought-V2 models chain-of-thought as a hidden-state trajectory and interleaves explicit text with continuous latent tokens, compressing steps whose transitions deviate from the question-to-solution direction. Trained via stepwise embedding forcing and label forcing with soft multi-modal supervision, it was evaluated on Qwen3.5-9B and Qwen3.6-27B across six benchmarks. Reported results include up to 2.6% average accuracy gain, up to 50% shorter responses, 2.29x higher Accuracy per Computation Unit, 94.6% faster preprocessing, and up to 80.3% faster training.

Hugging Face daily papers·Sep 7, 13:56 UTC · 9d agoAI research

Distill Globally, Adapt Locally: Reasoning Distillation and Product-Type Test-Time Training for Scalable Trade-Up Recommendation

A distillation framework compresses LLM reasoning into a 15.5M-parameter trade-up recommendation model reaching AUC 0.941 with product-type test-time training.

The paper targets trade-up recommendation, which identifies higher-quality alternatives that preserve customer purchase intent. A retrieval-augmented few-shot LLM teacher generates labels and rationales that supervise a compact embedding-pair classifier; at inference the 15.5M-parameter student uses only two precomputed 768-dimensional embeddings with no LLM calls. On 8,352 annotated pairs, label-only training scored AUC 0.912, reasoning distillation reached 0.924, and product-type test-time training lifted it to 0.941 with average precision 0.940. The distilled student is roughly 5,000x faster and 10,000x cheaper than direct LLM inference on a 100K-pair proxy catalog.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 17:08 UTC · 12d agoAI research1

A Zeroth-Order Paradigm for LLM Preference Alignment

ComPO is a zeroth-order preference alignment method using comparison oracles to mitigate likelihood displacement across Mistral, Llama, Gemma, and Qwen3 models.

The paper proposes Comparison-based Preference Optimization (ComPO), a zeroth-order alignment method that extracts directional information from preference pairs with small likelihood margins without directly optimizing a differentiable preference loss. The authors prove convergence guarantees for the offline scheme and performance guarantees for a constrained online variant with reverse-KL control. Experiments on Mistral, Llama, Gemma-2, Qwen3, and Gemma-3 show improvements over existing direct alignment methods, including length-controlled win rates, with pair-level diagnostics consistent with mitigating likelihood displacement.

Hugging Face daily papers·updated Sep 16, 17:59 UTC · 19h ago·first Sep 15, 20:00 UTC · 1d agoAI research 2 sources

JustFit: 200K-Token LLM Serving on a 24 GiB Laptop with Just-in-Time State Management

JustFit MLX runtime serves 200K-token contexts for Qwen3.8-27B on a 24 GiB MacBook via just-in-time state management.

JustFit is an MLX-based inference runtime combining KVExec for compressed KV execution, PhaseSwap for component residency, and StateTrans for state-preserving serving transitions, independent of weight quantization. On a 24 GiB M4 Pro MacBook running Qwen3.8-27B MXFP4, it completed 196,608 input and 16,384 output tokens, raising single-request context from the mlx-vlm baseline's 30,720 positions to 212,992 (6.93x). Performance tests show 19.11 tokens/s on a 32K-input probe with a 16,374 MiB median peak footprint, and the runtime answered 29 of 30 AIME 2026 problems correctly.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:15 UTC · 1d agoAI research

Context Engineering Inside the Harness: 4 Mechanisms That Beat Context Overflow and Goal Loss on Long-Horizon Tasks

Survey of four harness mechanisms—context budgeting, compaction, todo-state, and memory—that keep long-horizon LLM agents on task across 200+ tool calls.

The article details how agent harnesses, not larger context windows, solve context overflow and goal loss on long-horizon tasks, citing Chroma's Context Rot report showing 18 LLMs (GPT-4.1, Claude 4, Gemini 2.5, Qwen3) degrade on long inputs. Concrete implementations include LangChain Deep Agents offloading tool responses over 20,000 tokens to the filesystem and truncating old tool calls at 85% window usage, and Claude Code capping auto memory at 25KB while re-reading the 5 most recently modified files after compaction. OpenAI's Responses API now offers server-side compaction via context_management with a standalone /responses/compact endpoint, which Codex uses for long-running coding tasks. Manus reports a roughly 100:1 input-to-output token ratio per ~50-tool-call task, motivating todo.md state recitation to prevent goal drift.

MarkTechPost·Sep 13, 05:56 UTC · 4d agoAI research2

Google Research Releases ToolGrad: Answer-First Framework Hits 99.8% Pass Rate for Tool-Use Data Generation

Google Research and partners introduce ToolGrad, a verified tool-chain-first data generation framework reaching 99.8% pass rate and boosting Gemma-3-12B to 83.1 on BFCL.

Researchers from Google, the University of Tokyo, RIKEN AIP, and Tohoku University released ToolGrad, which inverts query-first tool-use data generation by executing and verifying API chains before annotating them with user queries. On the ToolBench database of 16,000+ APIs, ToolGrad raised generation pass rate from 63.8% to 99.8% while increasing tool uses per sample from 2.1 to 3.4 and cutting tool-use steps from 34.3 to 20.0. Fine-tuning Gemma-3 at 1B, 4B, and 12B parameters on the 500-sample ToolGrad-500 dataset lifted ToolGrad-12B to 83.1 on the Berkeley Function Calling Leaderboard, near Gemini 2.5 Pro at 83.2 and ahead of GPT-5 at 74.4. Code is Apache-2.0, with the dataset, PyPI package, and models available on Hugging Face.

MarkTechPost·Sep 11, 06:01 UTC · 6d agoAI research1

Training a 3.8B LLM to 0.384 CORE for $998 – Hugo Vergnes

Independent developer Hugo Vergnes trained a 3.8B-parameter Llama-style model to 0.384 CORE on 65B tokens for $998 in 43 hours on rented B200s.

Hugo Vergnes trained little-lm, a 3.848B-parameter decoder-only LLM, on 65.3B tokens in 43 hours for $998 using rented NVIDIA B200s, scoring 0.384 on the CORE benchmark and beating nanochat d32 (0.310) at similar cost. The Llama-style architecture uses RMSNorm, RoPE, GQA with 24 query and 8 KV heads, relu-squared MLPs, QK-norm, and ResFormer-style value embeddings that account for 19% of parameters. Key wins included the Muon optimizer for matrix parameters, a trapezoidal learning-rate schedule with linear cooldown, FP8 training plus vocabulary padding for roughly 33% throughput gains, and the ClimMix dataset over FineWeb-Edu. The project, inspired by Karpathy's nanochat, was built as a config-driven YAML framework for small LLM training.

Hacker News · security·Sep 10, 02:04 UTC · 7d agoAI researchHN 28↑ · 2 comments1

PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents

PARSER uses parallel reader subagents and an RL-trained lead agent for long-context QA, beating baselines and cutting latency up to 11x.

The PARSER paper decouples reading from reasoning: frozen subagents each read one document chunk in parallel while an RL-optimized lead agent iteratively broadcasts queries and aggregates evidence in scatter-gather rounds. On multi-hop QA with 7K to 896K token contexts, a 4B-backbone PARSER beats the strongest sequential memory baseline by 5.7 points on average and 12.0 points at 896K tokens, and a 9B version surpasses DeepSeek-V4-Pro by 6.3 points. Controlled experiments show robustness to evidence position, order, and distance perturbations, with inference latency reduced by up to 11x.

Hugging Face daily papers·Sep 5, 20:00 UTC · 11d agoAI research1