FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteJudgment incomplete136 / 162 · faca12f12723936d

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

Psychology and cognitive neuroscience · 2312.09257v2

FLOWING EVIDENCE BENCHMARK

How do we tell whether Evidence helps?

At the same writing position, with the same model and task, how does supplying retrieved paper passages change the first output? We compare matched versions and retain ties, unusable outputs and incomplete reviews.

SAME DRAFT · EVIDENCE ON OR OFF

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED INPUTS

SHARED BY BOTH

Manuscript context, model, task and prompt

A · With Evidence

Retrieved paper passages supplied

B · Without Evidence

No retrieved passages supplied

LLM

Same model and version

A → first output

B → first output

Blind judge agent

First outputs are anonymized as X and Y

Continuation review: accuracy, fit to the writing task and usability
Returns: X preferred / tie / Y preferred / both unusable

Order check: X / Y → Y / X

HOW DOES BLIND JUDGING WORK?

① Anonymize both outputs
The judge sees the same draft and both first outputs without knowing which received Evidence.

② Compare and swap order
The judge applies task-specific criteria in X/Y and then Y/X order.

③ Review disagreements
A third pass resolves disagreements. Incomplete reviews remain in the denominator.

How are positions and cards selected?

The evaluation uses seven fields with forty fixed ordinary autocomplete positions in each, for 280 total. A position does not need to require a literature citation in its next sentence. Each Flowing request receives three to five retrieved cards: five when all pass the relevance gate, without padding with weak cards. The run supplied 1,355 cards: 244 positions received five, 27 received four, and nine received three.

ADOPTION · 150/280

Adoption coverage

After drafting, the model returns an evidenceUse array identifying cards that materially affected facts, correction, direction, scope, terminology, or expression. A nonempty array counts as adoption.

OUTCOMES · 70 / 7 / 30

Full outcomes after adoption

Paired outcomes are counted only when the model reports adoption and both arms produce text. Among 107 comparable positions, Flowing was preferred in 70, 7 tied, and no-Evidence was preferred in 30; one tie was unusable in both arms.

How are the two table percentages calculated?

ADOPTION COVERAGE

150adopted
÷
280all positions
=
54%

FLOWING PREFERENCE AFTER ADOPTION

70Flowing preferred
÷
107comparable adopted positions
=
65%

evidenceUse is the model’s structured report of its own generation process. It does not require a citation and is not rejected merely because a faithful source paraphrase is not verbatim. Actual manuscript citations continue to use separate strict source verification. The previous 140-position result remains retained for traceability.

These are development-stage model-reported adoption and blind-judge results pending independent human review. They are not formal Benchmark conclusions and do not, on their own, establish causality.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

Brain-Inspired Machine Intelligence: A Survey of Neurobiologically-Plausible Credit Assignment

3 Discussion

…ormance gap between current state-of-the-art biological approaches to credit assignment and backprop-based DNNs [ 24 ] . Furthermore, more concerted effort will be needed to facilitate the construction of powerful, energy-efficient neuro-mimetic machines, as desired by the field. One of the domains that we argue might be most important will be the realm of RL/control, as this often encourages/lends itself to the design of modular neural agent architectures, often ones that share many similarities with the cognitive architectures typically crafted [ 11 , 224 , 402 , 315 ] in Cognitive Science.

…ed by the gradients computed by backprop [ 246 , 301 ] . Other results from this line of theoretical research include equivalences between bio-CA algorithms themselves [ 309 , 325 , 389 , 390 ] – for example, showing that predictive coding and contrastive Hebbian learning (and equilibrium propagation) can be equivalent to one another [ 283 ] or target propagation is (approximately) a form of Gauss-Newton optimization [ 31 , 276 ] . In [ 162 , 319 ] , relationships were established between forward-forward learning and contrastive Hebbian learning, contrastive divergence, and predictive coding.

↓ Next paragraph was generated here

The highlighted text is near the writing position. This is a layout of the evaluation input, not a PDF screenshot.

A matched comparison at the same writing positionJudgment: Judgment incomplete

Both arms use the same model, task and manuscript context. Only the retrieved passages differ. The first recorded output from each arm is shown below.

Evidence

Same model and position · retrieved passages supplied · first output

Generation status: failed. No complete output is available for this arm.

Passages supplied to the Evidence version

Constrained Predictive Coding as a Biologically Plausible Model of the Cortical Hierarchy

First are normative approaches, such as Predictive Coding [ 1 , 2 , 3 ] , Target Propagation and variations [ 44 , 45 , 47 ] , Equilibrium Propagation [ 49 , 53 ] and others [ 54 , 51 , 55 , 56 ] , where one starts from a mathematically motivated optimization objective. These methods, by virtue of t…

Read full passage excerpt

First are normative approaches, such as Predictive Coding [ 1 , 2 , 3 ] , Target Propagation and variations [ 44 , 45 , 47 ] , Equilibrium Propagation [ 49 , 53 ] and others [ 54 , 51 , 55 , 56 ] , where one starts from a mathematically motivated optimization objective. These methods, by virtue of their normative derivation, have a firm grounding in theory; however, they do not fully conform to the experimental observations in the brain (see below and Sec. 4 ). The second approach is driven by biology, with network structures and learning rules inspired by experimental findings [ 28 , 48 , 60 , 30 , 31 ] . While these works mostly conform to the experimentally observed findings, they are more challenging to analyze because of their conjectured phenomenological nature.

Model-reported card use

The model did not report that any card materially affected this completion.

No Evidence

Same model and position · no retrieved passages · first output

No continuation text was produced.

Model action：complete

Blind model judgment

The judgment was incomplete, so this case has no winner but remains in the denominator above.

The passages are input material; their presence does not mean an output used them or that they support every claim. Draft and source passages are excerpted; no simulated scores or PDF appear here.