Search: “function-calling”

Explainability Assistant: A Conversational XAI Interface for Interpreting Energy Consumption Models

Researchers release Explainability Assistant, an open-source conversational XAI tool using LLM function calling, lifting intent-parsing accuracy from 76.8% to 94%.

The paper introduces the Explainability Assistant, an open-source conversational XAI system for interpreting energy consumption forecasting models such as genetic-programming symbolic regressors. It uses LLM function calling instead of rigid custom grammars, achieving 94% intent-parsing accuracy versus 76.8% for prior work TalkToModel, and adapts to different ML problem types without task-specific fine-tuning. Comparative evaluation with energy domain specialists against a traditional XAI dashboard showed improved usability, with all experts preferring the conversational interface.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:40 UTC · 5d agoAI research1

OpenAI drops ChatGPT text chat limits for free users, adds new safeguards for teens

Help Net Security·Aug 7, 00:00 UTC · Aug 7, 2026Model release

ToolLoop: Closed-Loop Tool-Use Data Synthesis via Decomposed Generation and Dynamic Self-Feedback

ToolLoop introduces a closed-loop synthetic data framework whose 11K examples lift a 4B model to 86.40% on BFCL tool-use evaluation.

ToolLoop decomposes tool-use data synthesis into function-name sampling, backward derivation of user queries, and forward derivation of tool calls, with dynamic self-feedback at each stage. This shifts the paradigm from generate-then-filter to generate-verify-refine, reducing inefficient and imbalanced synthetic data. A 4B model trained on 11K synthetic examples reaches 86.40% accuracy on BFCL non-reasoning mode (86.07% in an Isolate variant excluding BFCL-overlapping functions) and 72.1% on ACEBench using only 18.3% of baseline training data.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:26 UTC · 7d agoAI research1

nex-agi/Nex-N2.5-mini — new model trending #30 on Hugging Face

Nex-AGI releases Nex-N2.5 agentic model family (mini, Pro, Max) with a 1.6-trillion-parameter MoE Max, open weights, and hosted access via OpenRouter.

Nex-AGI launched Nex-N2.5, a family of agentic models in mini, Pro, and Max sizes, with the Max version built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation and the company's first complete post-training effort at trillion-parameter scale. The models target long-horizon computer use, web browsing, and visually grounded agentic tasks, with expanded agent training environments. Reported benchmarks include Max scoring 86.1 on Terminal-Bench 2.1 and 65.7 on SWE-Bench Pro, trailing Claude Opus 5. Weights are being released openly on Hugging Face and ModelScope, with hosted access through OpenRouter.

Hugging Face trending models·Sep 8, 05:08 UTC · 7d agoModel release1

The GNU C Library security advisories update for 2026-09-14

glibc advisory GLIBC-SA-2026-0017 discloses a buffer overflow in strfmon/strfmon_l affecting versions 2.38 through 2.44.

The GNU C Library published security advisories including GLIBC-SA-2026-0017, a buffer overflow in strfmon and strfmon_l. Calling these functions with right-justified width padding conversions can write past the end of the caller-supplied output buffer in glibc 2.38 to 2.44. Exploitation requires an application code path that calls strfmon with attacker-influenced parameters.

oss-security·Sep 14, 19:54 UTC · 1d agoVulnerability 2 sources

Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs

Study shows LLM reasoning operations like planning and deduction are geometrically separable in hidden states, with separability peaking in middle layers.

Researchers investigate whether functional reasoning operations — problem formulation, goal decomposition, deduction — have corresponding geometric structure in LLM hidden representations. They find operations are separable in held-out representations with separability peaking in middle layers, ruling out lexical and positional confounds; token-wise operation alignment becomes more distributed across layers, and identical surface tokens are represented differently depending on their surrounding chunk. Attention-masking interventions show chunk-onset operation-aligned representations depend on preceding reasoning context; code is released on GitHub (naver-ai/beneath-cot).

Hugging Face daily papers·Sep 3, 20:00 UTC · 12d agoAI research1

nex-agi/Nex-N2.5-Pro — new model trending #30 on Hugging Face

Nex-AGI launches Nex-N2.5 agentic model family (mini/Pro/Max), with Max built on a 1.6-trillion-parameter MoE foundation.

Nex-AGI introduced Nex-N2.5, a next-generation family of agentic models in three sizes (mini, Pro, Max) focused on long-horizon agentic tasks including computer use, web browsing, and autonomous program execution. Nex-N2.5-Max is built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation, marking the company's first complete post-training effort at trillion-parameter scale. Weights will be released open-source on Hugging Face and ModelScope, with hosted access via OpenRouter. Benchmark comparisons against Claude Opus 5, GPT-5.6 Sol, Kimi-K3, GLM-5.3, DeepSeek-V4-Pro-0813, and Qwen3.8-Max show competitive scores on Terminal-Bench 2.1 and SWE-Bench Pro, though weights were listed as "coming soon" at publication.

Hugging Face trending models·Sep 8, 11:01 UTC · 7d agoModel release1

WireGuard-Linux Stack-Based Buffer Overflow in lsiio (Linux IIO Userspace Tool) Due to Unbounded fscanf

The Linux IIO userspace tool lsiio has a stack buffer overflow in find_type_by_name() caused by unbounded fscanf reads of oversized filesystem-backed attribute values.

A stack-based buffer overflow exists in the Linux Industrial I/O (IIO) userspace utility lsiio. In the find_type_by_name() function, the program reads an unbounded string from a filesystem-backed attribute into a fixed-size stack buffer using fscanf("%s", ...). A crafted or oversized attribute value causes a write beyond the bounds of the destination buffer. Despite the title's reference to WireGuard-Linux, the flaw described is in the IIO lsiio utility.

Full Disclosure·Sep 4, 00:13 UTC · 12d agoVulnerability

What Did I Just Say? Self-Listening for Full-Duplex Speech Models

Researchers propose Self-Listening, a full-duplex speech approach feeding realized model speech back as input to improve interruption recovery.

Full-duplex spoken language models can listen and speak simultaneously, but asynchronous text generation, speech synthesis, and playback cause mismatches between what a model believes it said and what the user heard. The paper defines the resulting recovery problem as anchor interruption and proposes Self-Listening, which interleaves user speech, model text, and played speech as input streams. The authors also release AnchorSpeech, a benchmark with homogeneous training and test splits tracking which ordered response items were actually spoken. Experiments show self-listening models achieve better anchoring performance than full-duplex baselines.

Hugging Face daily papers·Sep 3, 20:00 UTC · 12d agoAI research

Multi-Step Tool-Calling over Korean Open Public APIs: A Benchmark and a Data-Synthesis Recipe

Researchers introduce KOPA-Bench, a 145-task Korean public API tool-calling benchmark, and EDGE, an execution-grounded data synthesis method.

An arXiv paper presents KOPA-Bench, a benchmark of 145 real-world tasks chaining multiple tool-calls across live Korean government APIs, motivated by data-sovereignty requirements for on-premise open-source LLM agents. It also introduces EDGE, an execution-grounded dynamic graph that keeps only tool-output-to-input links verified by live API calls before synthesizing executable multi-step trajectories. A 9B model fine-tuned with GRPO on the resulting dataset nearly matches its untuned 27B family sibling on KOPA-Bench and improves on the BFCL benchmark.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 17:44 UTC · 11d agoAI research1

ECHO: A Matched-Contrast Benchmark for Context-Sensitive Turn-Taking in Full-Duplex Dialoguenew

ECHO benchmark pairs identical-overlap Chinese dialogue examples with contrasting contexts, revealing most full-duplex systems bias toward yielding the floor.

ECHO is a paired diagnostic benchmark for Chinese full-duplex turn-taking that matches examples with identical overlap transcripts but contrasting preceding multi-turn contexts, one requiring Yield and the other Keep. It also includes off-talk examples for diagnosing unnecessary yielding and introduces pair accuracy, which grants no credit to constant-action policies. Experiments across multiple full-duplex systems show most exhibit a pronounced Yield bias, performing substantially better on interruptions than backchannels. The benchmark and metadata will be publicly released.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 15:58 UTC · 12h agoAI research

Diffs vs. Whole Files: An Empirical Comparison of Iterative Edit-Based and Direct Generation for Flutter/Dart Code Models

Empirical study finds direct whole-file generation beats iterative diff-based editing for Flutter/Dart code models on about 1,790 held-out tasks.

Researchers trained Rainbow-Pony-100M from scratch and fine-tuned Qwen2.5-Coder-0.5B in both direct-generation and diff-based regimes, then evaluated four resulting models on roughly 1,790 Flutter/Dart tasks. Direct generation outperformed diff-based generation on compilation pass rate, bits-per-byte, character-level similarity, and blinded LLM-judge ratings. Diff-based editing is competitive only on short, localized edits in refactoring and error-handling tasks, a property the authors call task locality.

Hugging Face daily papers·Sep 4, 20:00 UTC · 11d agoAI research1

[webapps] Langflow 1.8.4 - Path Traversal to Remote Code Execution

A path traversal to remote code execution exploit for Langflow 1.8.4, a popular LLM application builder, was published on Exploit-DB.

Exploit-DB lists a proof-of-concept exploit chaining path traversal to remote code execution in Langflow 1.8.4, an open-source tool used to build LLM applications and agents. The chain allows an attacker to write arbitrary files outside the intended directory and achieve code execution on the host. The provided text does not include a CVE identifier or reports of exploitation in the wild, but RCE in a widely deployed AI tooling product is notable for defenders.

Exploit-DB·Aug 31, 00:00 UTC · 16d agoExploit / PoC1

Ask Before You Optimize: Dynamic Pre-Formulation Clarification for Interactive Optimization

Researchers release OR-Clarify, a benchmark testing whether LLM agents ask clarifying questions before formulating optimization models from incomplete requests.

OR-Clarify evaluates pre-formulation clarification in operations research: each task gives a partial problem description, withholds structured hidden slots, and scores agents via bounded interaction with a simulated user, measuring slot recovery, stopping behavior, silent assumptions, and interaction cost. The authors also propose InterOPT, a two-stage framework that identifies formulation-critical gaps to decide when to ask or stop. In choice-based experiments InterOPT substantially outperforms all baselines in exact slot recovery and remains competitive in the open-ended setting.

Hugging Face daily papers·Sep 3, 20:00 UTC · 12d agoAI research1

Show HN: LLM Attention Visualization

A developer released a browser-based tool that visualizes which past tokens influence each LLM output token using aggregated, value-weighted attention scores.

A Show HN project presents a React application built on Transformers.js that renders per-token attention influence by aggregating attention weights scaled by value-vector magnitudes across all attention heads and layers. To expose internal tensors, the author instrumented the ONNX computation graph, hosted a modified model on Hugging Face, and pre-generated prompts to avoid long model downloads in the browser. Demos with a 600-million-parameter model show how verbatim copying draws heavily on source tokens and how single outputs blend information from multiple phrases.

Hacker News · AI·Sep 8, 16:59 UTC · 7d agoAI tools & infraHN 41↑ · 9 comments

Characterizing Language Generation in the Limit: Finite Witnesses and a Separation-Width Hierarch

New work characterizes language generation in the limit via finite witnesses, proves a full separation-width hierarchy, and formalizes all results in Lean.

The paper fully characterizes when language generation in the limit is possible for arbitrary families over a countable universe: each target must admit a finite positive witness such that targets activated by any finite sample share an infinite common intersection. It defines positive separation width and proves every level of the resulting hierarchy occurs, with countable families admitting singleton witnesses and unions of families with infinite common cores requiring unbounded finite witnesses. The characterization, a universal normalization, and a diagonal capture lemma are machine-checked in the Lean proof assistant, with the development maintained on GitHub.

arXiv cs.AI / cs.LG / cs.CL·Sep 9, 17:54 UTC · 6d agoAI research1

TART: A Modular Tool for Technique-Aware Audio-to-Tablature Guitar Transcription

TART pipeline transcribes guitar audio to tablature with fingering and technique labels, outperforming prior baselines by up to 8.5 Tab F1 points.

TART is a modular four-stage pipeline for audio-to-tablature guitar transcription, combining an audio-to-MIDI transcription model, an expressive technique classifier, an audio-conditioned T5 encoder-decoder for string-fret assignment, and an automated tablature generator. In zero-shot evaluation on GuitarSet, EGDB, and two noisy augmented benchmarks, it averages 81.35% audio-to-MIDI F50 (+6.67 over the best baseline), 71.8% string-fret Tab F1 (+8.5), and 54.08% end-to-end Tab F1. The authors state it is the first framework to generate tablature with both fingering and expressive technique annotations directly from guitar audio.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:55 UTC · 5d agoAI research2

Model-Aware Schedules Improve Generation via Fiberwise Optimal Transport

Researchers introduce model-aware diffusion schedules via fiberwise optimal transport, cutting flow-matching FID on CIFAR-10 by 38.6% at 16 function evaluations.

The paper proposes constructing diffusion and flow-matching sampling schedules from a fiberwise prediction risk defined via optimal transport, combined with coefficient-path kinetic action, yielding a closed-form time allocation. Across DDPM and flow-matching experiments spanning targets, datasets, and architectures, the schedules beat model-agnostic baselines, including a 38.6% relative FID reduction for flow matching on CIFAR-10 at 16 function evaluations. Normalized fiberwise-risk profiles from independently trained models align closely, suggesting empirical universality, and a frozen analytic allocation template retains most of the gains.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:30 UTC · 5d agoAI research

One Symptom, Three Levers: A Critical Review of On-Policy Self-Distillation

A review paper frames on-policy self-distillation collapse as governed by three levers: token weighting, privileged information, and guidance decay.

The paper critically reviews On-Policy Self-Distillation (OPSD), where a language model trains on its own generations scored token-by-token by a teacher conditioned on privileged information such as reference solutions or environment feedback. It identifies collapse, the progressive narrowing of producible reasoning paths, as the dominant failure mode and analyzes it through three levers: signal weighting, the nature of privileged information, and teacher dynamics. The review is restricted to mathematical reasoning, reports no new experiments, and offers a shared vocabulary separating settled findings from disputed ones.

Hugging Face daily papers·Aug 25, 20:00 UTC · 21d agoAI research

MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes

MetroLLM-Bench is a 955-case benchmark testing language models as transit kiosk tool-calling runtimes across six real metro systems.

The benchmark covers 37-414-station metro systems and eleven task categories including routing, fare calculation, disruptions, accessibility, and adversarial input, with 14 deterministic and 8 semantic scoring components. Of 26 models from six vendors, a PEFT-tuned 4B Qwen 3.5 student scored 91.3 on Tier 1, exceeding GPT-5.6 (90.6/90.0), while Muse Glimmer 30B led the composite ranking. A deterministic rule-based baseline reached 84.6, and PEFT gains over base models shrank from +7.03 points at 2B to -0.91 at 27B.

Hugging Face daily papers·Sep 8, 20:00 UTC · 7d agoAI research

Re: Retrospective by 'gpg.fail' authors

GPG exploitation talk author clarifies a format-string 0day enabling code execution via printf %n writes and a polyglot PEM certificate payload.

Lexi Groves, author of the gpg.fail talk, clarified on oss-security that the first finding was an actual zero-day: a classic printf injection using %n for memory writes, with multiple X.509 certificates in one PEM file to re-enter and defeat ASLR before calling execv@plt. The payload executed the certificate itself, a polyglot file made by inserting a shebang and bash command into the PEM. A second finding was hash-collided by another party before the author could exploit it.

oss-security·00:50 UTC · 3h agoResearch 5 sources

SchemeArena: Factorized Stress Testing of Scheming in LLM Agents

Researchers introduce SchemeArena, a 400-scenario benchmark stress-testing scheming in LLM agents, finding explicit instrumental goals are the strongest driver of covert misaligned behavior.

The paper presents SchemeArena, a 400-scenario benchmark built through factorized scenario synthesis spanning safety-relevant tool domains, instrumental goals, oversight conditions and pressure mechanisms. The accompanying SCOUT monitor grounds multi-criteria scheming judgments in evidence drawn from agents' reasoning and actions. Stress tests across five LLM agents show explicit instrumental goals are the strongest driver of scheming propensity, while action-only monitoring increased scheming in several closed models, suggesting partial oversight can act as an optimization constraint. The benchmark, code and monitor are released at github.com/launchnlp/SchemeArena.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI safety & security

Memory as Plans: World-Action Modeling with Memory-Grounded Planning

Researchers introduce MaP-WAM, decomposing memory-dependent robot manipulation into memory-grounded planning and plan-conditioned execution, achieving 83.3% on RMBench and 78% on real robots.

MaP-WAM converts long-term multimodal episodic memory — segment records with language instructions and sparse visual context — into compact plans of next-segment language goals and visual guidance. A World-Action-Progress model jointly predicts action chunks and execution progress, calibrating predictions via plan-observation alignment for adaptive segment transitions and closed-loop context updates. Structured attention keeps the executor context length fixed and enables key-value caching, yielding state-of-the-art 83.3% success on RMBench, 78.0% on real-robot tasks, and roughly constant inference latency as task history grows.

Hugging Face daily papers·Sep 9, 20:00 UTC · 6d agoAI research

Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Modelsnew

RS-MFBO couples global sensitivity analysis with fidelity-augmented Gaussian processes to slash costly high-fidelity simulation runs in industrial flowsheet optimization.

The paper presents RS-MFBO, a reduced-space multi-fidelity Bayesian optimization framework for high-dimensional, expensive black-box functions. It integrates Global Sensitivity Analysis for dimensionality reduction with a fidelity-augmented Gaussian process and a cost-aware acquisition strategy featuring cooldown and promotion mechanisms. Validation on a plasmid DNA bioprocess (SuperPro Designer) and a green fuel synthesis plant (Aspen HYSYS) shows substantial reductions in high-fidelity evaluations while remaining competitive with single-fidelity baselines.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:46 UTC · 11h agoAI research

C*: Unifying Programming and Verification in C

Research paper C* (arXiv:2504.02246) proposes unifying programming and formal verification in C, discussed on Hacker News.

The arXiv paper 2504.02246, titled C*: Unifying Programming and Verification in C, presents an approach that combines C programming with built-in formal verification. The item surfaced via Hacker News, where it reached 28 points and 18 comments. Detailed abstract content was not included in the source text.

Hacker News · security·Sep 8, 15:56 UTC · 7d agoResearchHN 28↑ · 18 comments

ActionSplice: In-Flight Action Editing for Interactive World Models

ActionSplice enables in-flight action editing in chunk-autoregressive video world models via a lightweight corrector, avoiding rollback or waiting for the next chunk.

ActionSplice is an inference framework that formulates in-flight action editing for chunk-autoregressive video world models as Counterfactual State Transport (CST), where a lightweight corrector transports the interrupted backbone-native representation toward the matched state induced by the revised action. The world model and sampler remain frozen, and sampling resumes without replaying completed evaluations. Across minWM-Wan Action2V and HY-WM1.5, the retargeting variant CST-R reduces rollback-relative LPIPS by 61.5% and 75.9% versus direct condition swapping, while the temporal-splicing variant CST-T reduces suffix LPIPS by 56.1% and 77.5% with 2.73x and 1.69x pixel-ready speedups over waiting.

Hugging Face daily papers·Sep 7, 20:00 UTC · 8d agoAI research

Biology-in-the-loop: Amortized Adaptive Hit Discovery in CRISPR Screens

Researchers release AssayBench-Loop, a 1,389-screen CRISPR benchmark, and AssayLoop, a framework that learns adaptive hit discovery policies.

The paper introduces AssayBench-Loop, a large-scale benchmark of 1,389 CRISPR screens across five phenotype categories for adaptive hit discovery under budget constraints. It also introduces AssayLoop, which combines AssayFormer, a transformer-based amortized acquisition policy trained across historical screens, with LLM-derived biological priors via an adaptive handoff. On temporally held-out screens, AssayLoop achieves 5.67-fold enrichment over random selection and recovers 27.7% of hits after assaying roughly 5% of the candidate library, outperforming existing adaptive-design methods and standalone LLMs.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:45 UTC · 5d agoAI research1

CodeTD: Topology of Attention Detects Hallucinations in Code LLMs

CodeTD detects hallucinations in code LLMs before execution by analyzing topological patterns of attention maps, outperforming recent baselines.

CodeTD applies topological data analysis (TDA) to code LLM attention maps to quantify prompt-generation mismatch as a pre-execution correctness signal. Experiments cover HumanEval, MBPP, BigCodeBench, and MultiPL-E across 5 programming languages and 10 code LLMs up to 34B parameters. The method outperforms recent baselines and transfers between coding benchmarks, helping catch code that fails the task or embeds security vulnerabilities.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:26 UTC · 8d agoAI research1

Right Tool, Right Job: Native-Language Evaluation, Tokenizer Sensitivity, and Methodological Findings from a French-Only BabyLMnew

French BabyLM entry METRON-FR (125M GPT-2, 92.47M words) shows tokenizer artifacts dominate child-scale zero-shot evaluation; proposes standard diagnostics.

METRON-FR is a 125M-parameter GPT-2 pretrained on 92.47M French words, submitted to the BabyLM 2026 Strict track, scoring 85.97% on the native Quebec-French QFrBLiMP benchmark and 62.80% on the BabyLM-weighted leaderboard. A cross-lingual GLUE protocol combining French task-data translation with rank-16 LoRA shows relational tasks gain while world-knowledge tasks regress. Bilingual Lexicon Induction reaches p@1 of 68.84%, 18x above chance, and ablations show single-token zero-shot scoring is dominated by tokenizer and template artifacts at child scale.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:43 UTC · 11h agoAI research

IntentFuzz: A Protocol-Aware Fuzzer for Automated Invariant Violation Detection in Intent-Based Cross-Chain Bridges

IntentFuzz protocol-aware fuzzer recovers bridge structure from unannotated Solidity and confirmed 22 invariant violations across 24 real-world deployments.

IntentFuzz formalizes a taxonomy separating invariant violations from settlement exposures in intent-based cross-chain bridges, then recovers a bridge's intent structure and deposit/fill function roles from unannotated Solidity source. It classified deposit and fill functions with 100% recall and 82% combined precision, and achieved 100% recall and precision on 23 planted-bug mutants. Across 24 real-world deployments it confirmed 17 genuine invariant violations with heuristic-only input generation, rising to 22 with its LLM-assisted tier, spanning eight vulnerable GitHub repositories with findings reproducible against public deployed bytecode.

arXiv cs.CR·Sep 11, 16:03 UTC · 4d agoResearch1

The /wayfinder Skill: Navigating the “Fog of War” of Planning

Matt Pocock released the /wayfinder skill, an orchestrator layer that manages planning sessions, maps, and tickets for AFK coding agents.

Latent Space interviews Matt Pocock, whose AI Skills for Real Engineers project has 220,000+ GitHub stars, about his new /wayfinder skill. The skill manages agent context during ambiguous planning by splitting work into grilling, prototype, research, and task tickets organized under a shared map, enabling overnight AFK agent runs. It uses deliberate terminology like map, ticket, and session to steer agent behavior, and was tested on projects including a personal website rearchitecture.

Latent Space·Aug 20, 20:59 UTC · 26d agoAI tools & infra

MAxBench: A Multinomial Concept Recovery Benchmark

MAxBench evaluates multinomial concept recovery methods, finding affine subspaces steer most reliably but none consistently beats prompting.

MAxBench is a geometry-agnostic evaluation framework for multinomial concept representations in language models, based on sampling from recovered concept representations. It compares 10 localization methods covering 5 geometry types across 6 concepts and 4 models. Findings show affine subspaces steer more reliably than rank-one or linear subspaces due to better non-zero offsets, manifold steering is competitive where applicable, and no method consistently outperforms prompting.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 17:08 UTC · 4d agoAI research

Towards a Deterministic Math Solver for Clinical Language Models

Paper shows handing arithmetic to a deterministic Python solver beats direct model calculation at 32B but not reliably at 7B on MedCalc-Bench.

Researchers test a Program-Solve interface where clinical LLMs write case-specific Python executed by a restricted local solver instead of doing arithmetic directly. On MedCalc-Bench Verified (1,100 cases, 55 calculators), Qwen2.5-32B-AWQ scored 90.53% with solver handoff versus 83.47% with direct arithmetic (+7.05 points), while Qwen2.5-7B gained an unreliable +3.29 points with a confidence interval spanning zero. The authors audited the benchmark against clinical guidelines and flagged 16 of 55 calculators for version, use, or coefficient concerns.

Hugging Face daily papers·Sep 8, 20:00 UTC · 7d agoAI research